{"items":[{"id":"serverless-functions/cloudflare-workers","slug":"cloudflare-workers","category":"serverless-functions","access":"unconditional","accessBucket":"free","name":"Cloudflare Workers","provider":"Cloudflare Workers","planName":"freemium","kind":"Lightweight backend","description":"Best default for webhook receivers, setup APIs, and static frontend companion backend.","logo":"/logos/cloudflare-workers-paid.png","url":"https://developers.cloudflare.com/workers/","freeLimit":"Free tier historically centered around daily request quotas; verify current limits before launch.","bestFor":"Notion webhooks, Telegram webhooks, CORS APIs, auth boundaries.","caveats":"Long-running agent work may need queues, containers, or another worker.","sourceUrl":"https://developers.cloudflare.com/workers/platform/pricing/","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"daily","origin":"frontmatter"},{"key":"requiresCard","valueText":"False","origin":"frontmatter"}],"tags":["webhook","edge","serverless"],"pricingModel":"freemium","renewal":"daily","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"The free plan allows 100,000 inbound Worker invocations per day, resetting at 00:00 UTC.","label":"100k requests/day","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"requests/day","value":100000},{"detail":"Each Worker invocation is hard-capped at 10 milliseconds of CPU time on the free tier, compared to 30 seconds (default) on paid — making long-running or CPU-heavy work impossible.","label":"10ms CPU time per invocation","renewable":false,"sentiment":"neutral","severity":"high","type":"performance_throttled"},{"detail":"Free plan includes 100,000 KV read operations per day; exceeding this causes errors for the rest of the day.","label":"Workers KV: 100k reads/day","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"KV reads/day","value":100000},{"detail":"Free plan includes only 1,000 KV write operations per day, which is very tight for write-heavy workloads.","label":"Workers KV: 1k writes/day","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"KV writes/day","value":1000},{"detail":"Free plan caps total Workers KV stored data at 1 GB with no way to expand without upgrading.","label":"Workers KV: 1 GB storage","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"GB","value":1},{"detail":"Free D1 serverless SQL allows 5 million row reads per day, resetting at 00:00 UTC.","label":"D1: 5M rows read/day","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"D1 rows read/day","value":5000000},{"detail":"Free D1 allows only 100,000 row writes per day; write-heavy apps will hit this quickly.","label":"D1: 100k rows written/day","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"D1 rows written/day","value":100000},{"detail":"Total D1 database storage is capped at 5 GB across all databases on the free plan.","label":"D1: 5 GB total storage","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"GB","value":5},{"detail":"Free Queues includes 10,000 operations/day and enforces a non-configurable 24-hour message retention window.","label":"Queues: 10k ops/day, 24h retention","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"queue operations/day","value":10000},{"detail":"Durable Objects on the free plan allow 100,000 requests per day; paid plan scales to 1 million/month.","label":"Durable Objects: 100k requests/day","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"DO requests/day","value":100000},{"detail":"Free plan includes 200,000 log events written per day with only 3-day log retention vs 7 days on paid.","label":"Workers Logs: 200k events/day, 3-day retention","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"log events/day","value":200000},{"detail":"Free plan includes 100,000 Hyperdrive database proxy queries per day.","label":"Hyperdrive: 100k queries/day","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"Hyperdrive queries/day","value":100000},{"detail":"Free plan includes 30 million queried vector dimensions per month and 5 million stored vector dimensions.","label":"Vectorize: 30M queried vector dimensions/month","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"queried vector dimensions/month","value":30000000},{"detail":"Free R2 object storage includes 10 GB-month storage, 1 million Class A ops/month, and 10 million Class B ops/month with no egress charges.","label":"R2: 10 GB storage","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"GB","value":10},{"detail":"Workers Containers (for longer-running processes) are only available on the Workers Paid plan ($5/month minimum).","label":"No Containers on free tier","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Workers Trace Events Logpush (streaming logs to external destinations) is paid-only and unavailable on the free plan.","label":"No Logpush on free tier","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Free plan can only create SQLite-backed Durable Objects; key-value storage backend for Durable Objects requires the paid plan.","label":"No KV-backed Durable Objects on free","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"}],"keyMetrics":[{"display":"3.04M/mo","key":"invocations_per_month","label":"Invocations","note":"~3.04M/mo","primary":true,"unit":"M invocations/mo","value":3.04},{"display":"n/a","key":"compute_gb_seconds","label":"Compute","note":"duration not metered","unit":"K GB-seconds/mo"},{"display":"10 ms","key":"max_cpu_per_invocation","label":"CPU/call","unit":"ms","value":10},{"display":"∞","key":"bandwidth_per_month","label":"Bandwidth","note":"no egress charges","unit":"GB/mo"}],"verified":true,"timestamp":"2026-06-15T00:00:00+00:00","verifiedAt":"2026-06-15","badges":["🟢 Free forever","🟢 Commercial OK","🟠 Single resource"],"valueModel":{"bucket":"free","calc":{"allowance_usd_mo":0.61,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":5.8,"unit":"score","value":5.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":4.6,"unit":"$/mo","value":0.6080000000000001},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":6,"unit":"score","value":6},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8.3,"unit":"score","value":8.3}],"required_axis_count":3,"score":5.8,"score_formula":"weighted_components_v1","value":0.608,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":3040000}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/vercel-functions","slug":"vercel-functions","category":"serverless-functions","access":"unconditional","accessBucket":"free","name":"Vercel Functions","provider":"Vercel Functions","planName":"freemium","kind":"Serverless backend","description":"Frontend-friendly functions and deploy previews.","logo":"/logos/vercel-functions-fluid-compute-pro.png","url":"https://vercel.com/","freeLimit":"Hobby limits apply; function duration/request limits depend on runtime.","bestFor":"Simple web APIs for frontend apps.","caveats":"Check duration and background limitations before agents.","sourceUrl":"https://vercel.com/docs/functions","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"False","origin":"frontmatter"}],"tags":["serverless","frontend"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Hobby plan includes 1 million function invocations per month; cannot purchase additional usage on the free tier.","label":"1M invocations/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"invocations/month","value":1000000},{"detail":"Hobby plan includes only 4 hours of active CPU compute time per month for Vercel Functions under the Fluid compute model.","label":"4 hrs Active CPU/month","renewable":true,"sentiment":"pro","severity":"high","type":"quota_renewable","unit":"CPU-hours/month","value":4},{"detail":"Hobby plan includes 360 GB-hours of provisioned memory per month; exceeding this is not purchasable on Hobby.","label":"360 GB-hrs memory/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"GB-hours/month","value":360},{"detail":"Fast Data Transfer is capped at 100 GB per month on the Hobby plan.","label":"100 GB data transfer/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"GB/month","value":100},{"detail":"Hobby plan is limited to 12 bundled Vercel Functions per deployment when not using a framework that consolidates them.","label":"12 functions per deployment","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"functions","value":12},{"detail":"Hobby plan deploys functions to a single region (iad1 by default); multi-region deployment requires Pro or Enterprise.","label":"Single region only","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Hobby plan supports only one developer; team collaboration and additional seats require upgrading to Pro.","label":"Single developer seat","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"seat","value":1},{"detail":"Vercel explicitly states the Hobby plan is for personal, noncom use; production commercial apps must upgrade to Pro.","label":"Personal use only","renewable":false,"sentiment":"con","severity":"high","type":"no_commercial_use"},{"detail":"Production deployment functions are archived after 2 weeks of no invocations (preview deployments after 48 hours), adding at least 1 second to the first cold start after archival.","label":"Functions archived after inactivity","renewable":false,"sentiment":"neutral","severity":"medium","type":"cold_start_sleep"},{"detail":"Cold start prevention (keep-alive/warmed instances) is a Pro-only feature; Hobby functions always cold start after archival.","label":"No cold start prevention","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Hobby plan retains only 1 hour of runtime logs, making debugging production incidents very difficult.","label":"1-hour runtime log retention","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"hours","value":1},{"detail":"Build minutes (standard, enhanced, turbo, elastic machines) are not included on the Hobby plan and must be purchased; CI/CD builds consume metered build time.","label":"No build minutes included","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Hobby plan has no custom deployment environments; only production and preview environments are available.","label":"No custom environments","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Webhook-triggered deployments are not available on the Hobby plan.","label":"No webhook triggers for deployments","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"}],"keyMetrics":[{"display":"1M/mo","key":"invocations_per_month","label":"Invocations","primary":true,"unit":"M invocations/mo","value":1},{"display":"1.3M GB·s","key":"compute_gb_seconds","label":"Compute","note":"~1,296K GB-seconds","unit":"K GB-seconds/mo","value":1296},{"display":"1 min","key":"max_cpu_per_invocation","label":"CPU/call","note":"60,000 ms","unit":"ms","value":60000},{"display":"100 GB/mo","key":"bandwidth_per_month","label":"Bandwidth","unit":"GB/mo","value":100}],"verified":true,"timestamp":"2026-06-15T00:00:00+00:00","verifiedAt":"2026-06-15","badges":["🔴 Non-commercial","🟠 Cold start","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":30.8,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":7.1,"unit":"score","value":7.1,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":8,"unit":"$/mo","value":30.8000432},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":6.6,"unit":"score","value":6.6},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":5.5,"unit":"score","value":5.5}],"required_axis_count":3,"score":7.1,"score_formula":"weighted_components_v1","value":30.8,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"allowed_use":"noncom","source":"stated","value":false},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":1000000},{"key":"gb_second","source":"frontmatter_keyMetrics","unit":"K GB-seconds/mo","value":1296000},{"key":"bandwidth_gb","source":"frontmatter_keyMetrics","unit":"GB/mo","value":100}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/upstash-workflow","slug":"upstash-workflow","category":"serverless-functions","access":"unconditional","accessBucket":"free","name":"Upstash | Workflow","provider":"Upstash","planName":"freemium","kind":"Durable workflow engine","description":"Durable serverless workflow steps with retries, sleeps, callbacks, HTTP calls, and QStash-backed scheduling.","logo":"/logos/upstash.svg","url":"https://upstash.com/docs/workflow","freeLimit":"Free tier: 1,000 steps/day, 1,000 steps per workflow run, 10 concurrent steps, 1 MB max message size, 7 day max sleep, and 15 min max HTTP response duration.","bestFor":"Lightweight durable automations: read Notion, call an LLM, write an inbox item, send Telegram, record a run log.","caveats":"Every step and retry is metered. It is not a long-running agent container; use Upstash | Box when the job needs shell, git, filesystem state, or a coding-agent loop.","sourceUrl":"https://upstash.com/pricing/workflow","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"False","origin":"frontmatter"}],"highlights":["1,000 workflow steps/day free","Retries, sleeps, waits, calls, and invokes","Scheduled workflows use QStash schedules","Good fit for light agent-like automations"],"tags":["workflow","durable-execution","serverless","automation","qstash"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Free tier is capped at 1,000 workflow steps per day; each retry of a failed step counts as an additional billed step.","label":"1,000 steps/day","renewable":true,"sentiment":"pro","severity":"high","type":"quota_renewable","unit":"steps/day","value":1000},{"detail":"Only 10 steps can run in parallel at any time on the free tier; paid plans allow 100–1,000.","label":"10 concurrent steps max","renewable":true,"sentiment":"neutral","severity":"medium","type":"rate_limit","unit":"concurrent steps","value":10},{"detail":"A single workflow run cannot execute more than 1,000 steps total, regardless of plan.","label":"1,000 steps per workflow run","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"steps/run","value":1000},{"detail":"Step payloads and inter-step responses are capped at 1 MB; paid plans offer 10–50 MB.","label":"1 MB max message size","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"MB","value":1},{"detail":"Workflows can sleep for at most 7 days on the free tier; paid plans allow up to 1 year or unlimited.","label":"7-day max sleep duration","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"days","value":7},{"detail":"HTTP calls from a workflow step time out after 15 minutes; paid tiers allow 2–12 hours.","label":"15-minute max HTTP response duration","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"minutes","value":15},{"detail":"Dead-letter queue entries and execution logs are only retained for 3 days; paid plans extend this to 7–90 days.","label":"3-day DLQ and log retention","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"days","value":3},{"detail":"The free tier carries no uptime SLA; SLA tiers are only available on paid Fixed and Enterprise plans.","label":"No uptime SLA","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Once the 1,000 steps/day quota is hit, QStash returns errors and the SDK throttles retries; runs that exhaust retries are failed and moved to the DLQ.","label":"Step execution throttled on daily limit","renewable":true,"sentiment":"neutral","severity":"high","type":"performance_throttled"},{"detail":"Free tier includes 50 GB/month of bandwidth; usage beyond this is billed at $0.05/GB.","label":"50 GB bandwidth/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"GB/month","value":50}],"keyMetrics":[{"display":"0.03M/mo","key":"invocations_per_month","label":"Invocations","note":"~30K steps/mo","primary":true,"unit":"M invocations/mo","value":0.03},{"display":"n/a","key":"compute_gb_seconds","label":"Compute","note":"billed per step, not GB-s","unit":"K GB-seconds/mo"},{"display":"n/a","key":"max_cpu_per_invocation","label":"CPU/call","note":"no CPU cap; 15-min max HTTP response","unit":"ms"},{"display":"50 GB/mo","key":"bandwidth_per_month","label":"Bandwidth","unit":"GB/mo","value":50}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","verifiedAt":"2026-06-17","badges":["🟢 Commercial OK","🟠 Single resource"],"auditNote":"no FPS in parallel report — used audit recommendation","valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":4.51,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":6.2,"unit":"score","value":6.2,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":6.3,"unit":"$/mo","value":4.506},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":4.7,"unit":"score","value":4.7},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8.3,"unit":"score","value":8.3}],"required_axis_count":3,"score":6.2,"score_formula":"weighted_components_v1","value":4.506,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":30000},{"key":"bandwidth_gb","source":"frontmatter_keyMetrics","unit":"GB/mo","value":50}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/deno-deploy","slug":"deno-deploy","category":"serverless-functions","access":"unconditional","accessBucket":"free","name":"Deno Deploy","provider":"Deno Deploy","planName":"freemium","kind":"Edge JavaScript/TypeScript serverless runtime","description":"Deno Deploy runs JavaScript/TypeScript at the edge with a built-in KV store and a free tier for personal projects.","logo":"/logos/deno-deploy-pro.png","url":"https://deno.com/deploy/pricing","freeLimit":"Free plan: 1M inbound requests/month, 20 GB egress, 1 GiB volume storage, 1 GiB Deno KV storage, 450k KV read units and 300k KV write units per month, 50 custom domains.","bestFor":"Edge functions and small full-stack TS apps that want a built-in KV database without managing infrastructure.","caveats":"Per-request CPU time is limited; KV read/write units are metered; egress capped at 20 GB on free.","sourceUrl":"https://deno.com/deploy/pricing","offerFacts":[{"key":"offered_price_usd","valueNumber":20,"valueText":"20.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"False","origin":"frontmatter"}],"highlights":["1M requests/month free","1 GiB built-in Deno KV storage included","20 GB egress and 50 custom domains free","First-party edge runtime by the Deno team"],"tags":["edge","serverless","typescript","kv","functions"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Free plan allows up to 1 million inbound HTTP requests per month across all deployments; resets each billing cycle.","label":"1M requests/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"requests/month","value":1000000},{"detail":"Monthly outbound bandwidth is capped at 20 GB across all deployments; inbound data does not count against this.","label":"20 GB egress/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"GB/month","value":20},{"detail":"Total CPU execution time across all deployments is limited to 15 hours per month on the new runtime (does not apply to Deno Deploy Classic).","label":"15h CPU time/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"CPU-hours/month","value":15},{"detail":"Total memory-time budget is 350 GB-hours per month; this applies to the new runtime only.","label":"350 GB-h memory time/month","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"GB-h/month","value":350},{"detail":"Deno KV read operations (get, list) are limited to 450,000 units (4 KiB each) per month.","label":"450k KV read units/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"KV read units/month","value":450000},{"detail":"Deno KV write operations (set, delete) are limited to 300,000 units (1 KiB each) per month.","label":"300k KV write units/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"KV write units/month","value":300000},{"detail":"The free plan allows at most 60 new deployments per hour, shared across the organization.","label":"60 deployments/hour","renewable":true,"sentiment":"neutral","severity":"low","type":"rate_limit","unit":"deployments/hour","value":60},{"detail":"Free plan is locked to a single database region with no ability to select the write region, limiting global-latency use cases.","label":"1 KV DB region","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Wildcard subdomain routing is not available on the free plan; Pro and above support it.","label":"No wildcard subdomains","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Logs are retained for only 1 day on the free plan; Pro gets 1 week and Builder gets 2 weeks.","label":"1-day log retention","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Free plan is limited to community (Discord) support only; no email support, support SLA, or reliability SLA is provided.","label":"No email support or SLA","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"The ability to require authentication credentials for all incoming requests to a deployment is an Enterprise-only feature.","label":"No authenticated invocations","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"SOC2 Type 1 certification and Data Processing Agreements are Enterprise-only, making the free tier unsuitable for regulated workloads.","label":"No SOC2 / DPA compliance","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Only 20 deployments can be active at any given time within a free organization.","label":"Max 20 active apps","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"apps","value":20},{"detail":"Free organizations are limited to 5 authenticated team members who can manage resources.","label":"5 team members","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"members","value":5},{"detail":"Deno explicitly positions the free plan as 'for personal use and smaller projects', signaling it is not intended for production commercial workloads.","label":"Personal use framing","renewable":false,"sentiment":"neutral","severity":"info","type":"feature_restricted"}],"keyMetrics":[{"display":"1M/mo","key":"invocations_per_month","label":"Invocations","primary":true,"unit":"M invocations/mo","value":1},{"display":"1.26M GB·s","key":"compute_gb_seconds","label":"Compute","note":"~1,260K GB-s","unit":"K GB-seconds/mo","value":1260},{"display":"n/a","key":"max_cpu_per_invocation","label":"CPU/call","note":"15h CPU/mo org-wide","unit":"ms"},{"display":"20 GB/mo","key":"bandwidth_per_month","label":"Bandwidth","unit":"GB/mo","value":20}],"verified":true,"timestamp":"2026-06-16T00:00:00+00:00","verifiedAt":"2026-06-16","badges":["🟢 Commercial OK","🟠 Single resource"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":23,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":7.2,"unit":"score","value":7.2,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":7.7,"unit":"$/mo","value":23.000042},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":5.7,"unit":"score","value":5.7},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8.3,"unit":"score","value":8.3}],"required_axis_count":3,"score":7.2,"score_formula":"weighted_components_v1","value":23,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":1000000},{"key":"gb_second","source":"frontmatter_keyMetrics","unit":"K GB-seconds/mo","value":1260000},{"key":"bandwidth_gb","source":"frontmatter_keyMetrics","unit":"GB/mo","value":20}],"raw":{"offered_price_usd":20,"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/val-town","slug":"val-town","category":"serverless-functions","access":"unconditional","accessBucket":"free","name":"Val Town","provider":"Val Town","planName":"freemium","kind":"Hosted serverless TypeScript/JavaScript functions ('vals')","description":"Val Town lets you write, run and schedule small TypeScript/JavaScript functions in the cloud with HTTP endpoints, crons and an MCP integration.","logo":"/logos/val-town.png","url":"https://www.val.town/pricing","freeLimit":"Free plan: 100,000 runs/day, unlimited public vals, 15-minute cron intervals, 1-minute wall-clock time per run, 3-day log retention, MCP integration. (Free tier allows 0 private vals.)","bestFor":"Quick serverless scripts, webhooks, cron jobs and AI/agent tool endpoints without any deployment setup.","caveats":"Free tier vals are public only (no private/unlisted); short per-run wall-clock limit; no custom domains on free.","sourceUrl":"https://www.val.town/pricing","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"False","origin":"frontmatter"}],"highlights":["100,000 runs per day on the free tier","Unlimited public vals with HTTP endpoints","Built-in cron scheduling (15-min intervals)","MCP integration for agent tools"],"tags":["serverless","typescript","cron","webhooks","mcp"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"The free plan allows up to 100,000 val executions per day; this quota resets every 24 hours.","label":"100,000 runs/day","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"runs/day","value":100000},{"detail":"Free tier vals must be public — zero private or unlisted vals are permitted; private vals require a paid plan.","label":"No private vals","renewable":false,"sentiment":"neutral","severity":"high","type":"feature_restricted"},{"detail":"Custom domain mapping is not available on the free plan; vals are served under the val.town subdomain only.","label":"No custom domains","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Scheduled vals can run no more frequently than every 15 minutes; 1-minute intervals require Pro.","label":"15-minute minimum cron interval","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"minutes","value":15},{"detail":"Each val execution is hard-limited to 1 minute of wall-clock time, making long-running tasks impossible on the free tier.","label":"1-minute wall-clock time per run","renewable":false,"sentiment":"neutral","severity":"medium","type":"performance_throttled","unit":"minutes/run","value":1},{"detail":"Execution logs are kept for only 3 days on the free plan; Pro extends this to 10 days.","label":"3-day log retention","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"days","value":3},{"detail":"The free plan restricts email sending to the account owner's own address; emailing arbitrary recipients requires Pro.","label":"Email only to self","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"The built-in AI coding assistant (Townie) credit is not included on the free plan; Pro includes $10/month.","label":"No Townie (AI) credit","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"}],"keyMetrics":[{"display":"3.04M/mo","key":"invocations_per_month","label":"Invocations","note":"100K runs/day","primary":true,"unit":"M invocations/mo","value":3.04},{"display":"∞","key":"compute_gb_seconds","label":"Compute","unit":"K GB-seconds/mo"},{"display":"1 min","key":"max_cpu_per_invocation","label":"CPU/call","note":"1 min wall-clock/run","unit":"ms","value":60000},{"display":"∞","key":"bandwidth_per_month","label":"Bandwidth","unit":"GB/mo"}],"verified":true,"timestamp":"2026-06-16T00:00:00+00:00","verifiedAt":"2026-06-16","badges":["🟢 Commercial OK","🟢 No signup"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":0.61,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":6,"unit":"score","value":6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":4.6,"unit":"$/mo","value":0.6080000000000001},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":6.2,"unit":"score","value":6.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9}],"required_axis_count":3,"score":6,"score_formula":"weighted_components_v1","value":0.608,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":3040000}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/modal","slug":"modal","category":"serverless-functions","access":"unconditional","accessBucket":"free","name":"Modal","provider":"Modal","planName":"freemium","kind":"Serverless compute / GPU runtime","description":"Serverless compute platform for running Python functions, containers, and GPU workloads with a recurring $30/month free compute credit on the Starter plan.","logo":"/logos/modal.png","url":"https://modal.com/pricing","freeLimit":"Starter plan ($0): $30/month in free compute credits (recurring), 3 workspace seats, 100 containers + 10 GPU concurrency, 5 deployed crons, 8 webhook endpoints, 200 deployed apps, 1 TiB/month free volume storage.","bestFor":"Running serverless Python jobs, scheduled tasks, web endpoints, and GPU/ML inference where you want recurring monthly compute credit rather than a one-time trial.","caveats":"The $30 is a monthly compute credit, not unlimited free usage; heavy GPU jobs exhaust it quickly. Log retention on Starter is 1 day; custom domains and static IP require the Team plan.","sourceUrl":"https://modal.com/pricing","offerFacts":[{"key":"creditAmountUsd","valueNumber":30,"valueText":"30","origin":"frontmatter"},{"key":"credit_amount_usd","valueNumber":30,"valueText":"30.0","origin":"value_model_raw"},{"key":"offered_price_usd","valueNumber":30,"valueText":"30.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"False","origin":"frontmatter"}],"highlights":["$30/month recurring free compute credit, not a one-time trial","100 containers and 10 GPU concurrency on the free Starter plan","1 TiB/month of free volume storage included","Pay-per-second billing with scale-to-zero, no idle charges"],"tags":["serverless","gpu","python","compute","ml-inference"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":30,"eligibility":["anyone"],"limitations":[{"detail":"Starter plan includes $30 in compute credits that reset each calendar month; unused credits do not roll over.","label":"$30/month compute credit","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"USD/month","value":30},{"detail":"Volume storage is free up to 1 TiB per month; overages are billed at $0.09/GiB/month.","label":"1 TiB/month free volume storage","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"GiB/month","value":1024},{"detail":"The Starter plan is capped at 3 workspace seats; unlimited seats require the Team plan ($250/month).","label":"3 workspace seats max","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"seats","value":3},{"detail":"Concurrent container count is capped at 100 and GPU concurrency at 10; Team plan raises these to 1000 and 50 respectively.","label":"100 containers / 10 GPU concurrency","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"containers","value":100},{"detail":"Only 5 cron (scheduled) jobs can be deployed simultaneously on the Starter plan; Team plan offers unlimited.","label":"5 deployed cron jobs","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"crons","value":5},{"detail":"Webhook (web function) deployments are capped at 8 endpoints on Starter; Team plan is unlimited.","label":"8 webhook endpoints","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"endpoints","value":8},{"detail":"Maximum of 200 deployed apps on the Starter plan; Team plan allows up to 1000.","label":"200 deployed apps","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"apps","value":200},{"detail":"Logs are retained for only 1 day on Starter; Team plan provides 30-day retention and Enterprise allows custom retention.","label":"1-day log retention","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"days","value":1},{"detail":"Custom domains for web endpoints are not available on the Starter plan and require the Team plan.","label":"No custom domains","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Static IP egress proxy is a Team/Enterprise-only feature; Starter traffic uses shared IPs.","label":"No static IP proxy","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Deployment rollback history (3 versions) is only available on the Team plan; Starter has no rollback capability.","label":"No deployment rollbacks","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"HIPAA compatibility, Okta SSO, and audit logs are Enterprise-only features unavailable on the free Starter plan.","label":"No HIPAA / SSO / audit logs","renewable":false,"sentiment":"neutral","severity":"info","type":"feature_restricted"}],"keyMetrics":[{"display":"n/a","key":"invocations_per_month","label":"Invocations","note":"$30/mo credit-gated","primary":true,"unit":"M invocations/mo"},{"display":"n/a","key":"compute_gb_seconds","label":"Compute","note":"$30/mo credit; CPU $0.0000131/core-s + mem $0.00000222/GiB-s","unit":"K GB-seconds/mo"},{"display":"Uncapped","key":"max_cpu_per_invocation","label":"CPU/call","note":"limited by 100 containers / 10 GPU concurrency","unit":"ms"},{"display":"∞","key":"bandwidth_per_month","label":"Bandwidth","note":"no egress fees","unit":"GB/mo"}],"verified":true,"timestamp":"2026-06-16T00:00:00+00:00","verifiedAt":"2026-06-16","badges":["🟢 Commercial OK","🔴 High minimum/overage"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":5.8,"unit":"score","value":5.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":1,"b":1,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_credit","score":4,"unit":"$ one-time","value":30},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":7.2,"unit":"score","value":7.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8.3,"unit":"score","value":8.3}],"required_axis_count":3,"score":5.8,"score_formula":"weighted_components_v1","value":30,"value_kind":"one_time_usd","value_provenance":"provider_credit"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"credit_amount_usd":30,"offered_price_usd":30,"pricingModel":"freemium"},"subtop":"credits"}},{"id":"serverless-functions/github-actions","slug":"github-actions","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"GitHub Actions","provider":"GitHub Actions","planName":"freemium","kind":"Scheduled worker","description":"Free-ish scheduled or manual jobs for technical users.","logo":"/logos/github.svg","url":"https://github.com/features/actions","freeLimit":"Free minutes for public repos and limited private-repo minutes depending on account.","bestFor":"Daily/hourly exports, repo automation, batch processing.","caveats":"Not an always-on webhook server.","sourceUrl":"https://docs.github.com/actions","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"False","origin":"frontmatter"}],"tags":["ci","scheduled","technical"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"GitHub Free accounts get 2,000 GitHub-hosted runner minutes per month for private repositories; resets at the start of each billing cycle.","label":"2,000 free minutes/month (Free plan)","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"minutes/month","value":2000},{"detail":"GitHub Free accounts include 500 MB of artifact and Packages storage shared across Actions and GitHub Packages; resets monthly.","label":"500 MB artifact storage (Free plan)","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"MB","value":500},{"detail":"Each repository gets 10 GB of Actions cache storage included; usage beyond this is billed at $0.07/GiB/month.","label":"10 GB cache storage per repo","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"GB","value":10},{"detail":"Minutes are unlimited and free for public repositories using standard runners; the 2,000-minute cap only applies to private repositories.","label":"Private repo minutes are capped; public repos are free","renewable":false,"sentiment":"neutral","severity":"info","type":"feature_restricted"},{"detail":"GitHub-hosted larger (more powerful) runners are always charged even in public repositories or when free quota is available.","label":"Larger runners always billed","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Windows runner minutes cost ~1.67x and macOS runner minutes cost ~10x the Linux rate against the free quota, shrinking effective free compute significantly.","label":"macOS/Windows runners consume minutes faster","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"GitHub Actions is event/schedule-driven and cannot act as a persistent webhook listener or always-on HTTP server.","label":"Not an always-on server","renewable":false,"sentiment":"neutral","severity":"high","type":"feature_restricted"},{"detail":"Jobs on free-tier GitHub-hosted runners are queued and may wait before execution; there is no guaranteed instant cold-start.","label":"Job queue delay on shared runners","renewable":false,"sentiment":"neutral","severity":"low","type":"cold_start_sleep"},{"detail":"Free standard runners use shared 2-core GitHub-hosted machines; performance is not dedicated and can vary under load.","label":"Shared runner hardware (standard tier)","renewable":false,"sentiment":"neutral","severity":"low","type":"performance_throttled"},{"detail":"If no valid payment method is on file, all Actions usage is blocked once the monthly quota is exhausted; no grace period.","label":"Card required to use quota beyond free tier","renewable":false,"sentiment":"con","severity":"medium","type":"requires_card"},{"detail":"The 500 MB free storage allowance is pooled across Actions artifacts, Actions caches, and GitHub Packages — heavy Packages use eats into Actions storage.","label":"Artifact storage shared with GitHub Packages","renewable":false,"sentiment":"con","severity":"medium","type":"total_volume_cap"}],"keyMetrics":[{"display":"n/a","key":"invocations_per_month","label":"Invocations","note":"billed by minutes, not invocations","primary":true,"unit":"M invocations/mo"},{"display":"∞","key":"compute_gb_seconds","label":"Compute","note":"private","unit":"K GB-seconds/mo"},{"display":"n/a","key":"max_cpu_per_invocation","label":"CPU/call","note":"6h job / 35-day workflow limits","unit":"ms"},{"display":"n/a","key":"shared_bandwidth_gb","label":"Bandwidth","note":"shared w/ Packages","unit":"GB/mo"}],"verified":true,"timestamp":"2026-06-15T00:00:00+00:00","verifiedAt":"2026-06-15","badges":["🟢 Commercial OK","🟠 Cold start","🔴 Card/balance required","🔴 One-time","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":3.2,"unit":"score","value":3.2,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$/mo","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":4.8,"unit":"score","value":4.8},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":6.3,"unit":"score","value":6.3}],"required_axis_count":3,"score":3.2,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"usd_month","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/netlify-functions","slug":"netlify-functions","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Netlify Functions","provider":"Netlify Functions","planName":"freemium","kind":"Serverless backend","description":"Static hosting plus functions for simple APIs.","logo":"/logos/netlify.svg","url":"https://www.netlify.com/","freeLimit":"Free tier exists; function/runtime quotas depend on current plan.","bestFor":"Small API endpoints next to static frontend.","caveats":"Not a full always-on agent runtime.","sourceUrl":"https://docs.netlify.com/functions/overview/","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"False","origin":"frontmatter"}],"tags":["serverless","static"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Free tier gets 300 Netlify credits per month covering all usage (compute, deploys, bandwidth, requests); the cap is hard — sites pause when exhausted.","label":"300 credits/month (hard cap)","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"credits/month","value":300},{"detail":"Function and edge function compute is billed at 10 credits/GB-hour, so the 300-credit pool yields roughly 30 GB-hours of serverless function runtime per cycle.","label":"~30 GB-hours compute/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"GB-hours/month","value":30},{"detail":"Web requests (including function invocations) consume 2 credits per 10k requests; using the full credit pool for requests alone gives about 1.5 million requests.","label":"~1.5M web requests/month","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"requests/month","value":1500000},{"detail":"Bandwidth is charged at 20 credits/GB; spending the full 300 credits on bandwidth would yield 15 GB, but in practice credits are shared across all usage types.","label":"~15 GB bandwidth/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"GB/month","value":15},{"detail":"Each production deploy costs 15 credits, so the 300-credit pool supports approximately 20 production deploys per month if used exclusively for deploys.","label":"~20 production deploys/month","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"deploys/month","value":20},{"detail":"Function and edge function logs are retained for only 24 hours on the free plan, making debugging production issues significantly harder.","label":"24-hour log retention only","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Analytics, metrics, and observability data are limited to 1-day history on the free tier, versus 7-30 days on paid plans.","label":"1-day observability history","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"The free plan is limited to 1 member seat; additional paid seats cannot be added on the free plan.","label":"Single team member","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted","unit":"members","value":1},{"detail":"Increased execution limits, custom memory, and custom AWS region for functions are paid-tier features only.","label":"No custom function deployment options","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Free tier allows only 1 concurrent build; additional concurrent builds require paid add-ons at $40 each.","label":"1 concurrent build","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"concurrent builds","value":1},{"detail":"Free plan includes only the 'Basic' tier of the built-in Postgres database; relaxed and advanced tiers require paid plans.","label":"Basic Netlify Database only","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"When the 300-credit monthly limit is hit, ALL projects on the account are paused until the next billing cycle or a plan upgrade — there is no graceful degradation.","label":"Site pauses when credits exhausted","renewable":false,"sentiment":"con","severity":"high","type":"total_volume_cap"},{"detail":"Netlify Functions are event-driven serverless functions with cold starts; they are not suitable as persistent long-running agent or background runtimes.","label":"Not an always-on runtime","renewable":false,"sentiment":"neutral","severity":"medium","type":"cold_start_sleep"}],"keyMetrics":[{"display":"1.5M/mo","key":"shared_invocations_per_month","label":"Invocations","note":"300 credits, shared","primary":true,"unit":"M invocations/mo","value":1.5},{"display":"108K GB·s","key":"shared_compute_gb_seconds","label":"Compute","note":"30 GB-hr, shared","unit":"K GB-seconds/mo","value":108},{"display":"10 s","key":"max_cpu_per_invocation","label":"CPU/call","unit":"ms","value":10000},{"display":"15 GB/mo","key":"shared_bandwidth_gb","label":"Bandwidth","note":"300 credits, shared","unit":"GB/mo","value":15}],"verified":true,"timestamp":"2026-06-15T00:00:00+00:00","verifiedAt":"2026-06-15","badges":["🟢 Commercial OK","🔴 One-time","🟠 Cold start","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":3.8,"unit":"score","value":3.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$/mo","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":5.8,"unit":"score","value":5.8},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":7.6,"unit":"score","value":7.6}],"required_axis_count":3,"score":3.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"usd_month","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/aws-lambda","slug":"aws-lambda","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"AWS Lambda","provider":"AWS Lambda","planName":"freemium","kind":"Serverless Functions (FaaS)","description":"Event-driven serverless compute that runs your code on demand without provisioning or managing servers, billed per request and per millisecond of compute.","logo":"/logos/aws-lambda.png","url":"https://aws.amazon.com/lambda/","freeLimit":"1,000,000 requests and 400,000 GB-seconds of compute per month, free forever (shared across x86 and Graviton2/Arm), plus 100 GiB of HTTP response streaming per month.","bestFor":"Event-driven APIs, webhooks, cron jobs, and low-to-moderate-traffic backends that fit within the perpetual monthly allowance.","caveats":"The free allowance never expires, but creating an AWS account requires a payment method on file, and exceeding the monthly quota silently bills at standard pay-as-you-go rates. Provisioned Concurrency is excluded from the free tier and billed from the first GB-second. Each invocation is capped at 15 minutes and 10,240 MB of memory, and accounts start with a default 1,000 concurrent-execution limit.","sourceUrl":"https://aws.amazon.com/lambda/pricing/","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"True","origin":"frontmatter"}],"highlights":["Perpetual always-free tier: 1M requests + 400K GB-seconds every month, never expires","Works on both x86 and Graviton2/Arm processors against the same aggregate quota","Includes 100 GiB/month of HTTP response streaming (first 6 MB per request always free)","Per-millisecond billing once over quota; pay only for actual compute used"],"tags":["serverless","faas","functions","aws","compute","event-driven","always-free"],"pricingModel":"freemium","renewal":"monthly","requiresCard":true,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"1,000,000 requests and 400,000 GB-seconds of compute per month, resetting each calendar month. Usage beyond this bills at standard pay-as-you-go rates.","label":"Monthly free quota","renewable":true,"sentiment":"pro","severity":"info","type":"quota_renewable","unit":"requests/mo","value":1000000},{"detail":"Creating an AWS account requires a valid credit/debit card on file, even when staying entirely within the free tier.","label":"Payment method required at signup","renewable":false,"sentiment":"con","severity":"medium","type":"requires_card"},{"detail":"The free tier does not apply to functions with Provisioned Concurrency enabled; those are billed for Requests and Duration from the first unit.","label":"Provisioned Concurrency excluded","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"A single invocation can run for at most 900 seconds (15 minutes); longer-running work is terminated with a timeout.","label":"15-minute invocation timeout","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"seconds","value":900},{"detail":"Function memory is configurable from 128 MB up to 10,240 MB (with proportional vCPU); the same allocation also caps available CPU.","label":"Memory cap per function","renewable":false,"sentiment":"neutral","severity":"info","type":"single_resource_cap","unit":"MB","value":10240},{"detail":"Accounts start with a default soft limit of 1,000 concurrent executions per region; requests beyond this are throttled until a quota increase is granted.","label":"Default concurrency limit","renewable":false,"sentiment":"neutral","severity":"medium","type":"rate_limit","unit":"concurrent executions","value":1000},{"detail":"100 GiB of HTTP response streaming per month is included free (beyond the first 6 MB per request, which is always free); excess streamed bytes are billed at $0.008/GB.","label":"Response streaming cap","renewable":true,"sentiment":"neutral","severity":"info","type":"total_volume_cap","unit":"GiB/mo","value":100},{"detail":"Without (paid) Provisioned Concurrency, infrequently invoked functions incur cold-start latency while a new execution environment is initialized.","label":"Cold starts on idle functions","renewable":false,"sentiment":"neutral","severity":"low","type":"cold_start_sleep"}],"keyMetrics":[{"display":"1M/mo","key":"invocations_per_month","label":"Invocations","primary":true,"unit":"M invocations/mo","value":1},{"display":"400K GB·s","key":"compute_gb_seconds","label":"Compute","unit":"K GB-seconds/mo","value":400},{"display":"15 min","key":"max_cpu_per_invocation","label":"CPU/call","note":"15 min","unit":"ms","value":900000},{"display":"100 GB/mo","key":"bandwidth_per_month","label":"Bandwidth","unit":"GB/mo","value":100}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","verifiedAt":"2026-06-17","badges":["🟢 Commercial OK","🔴 Card/balance required","🟠 Single resource","🔴 One-time","🟠 Cold start","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":15.87,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":6.2,"unit":"score","value":6.2,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":7.4,"unit":"$/mo","value":15.86668},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":4.3,"unit":"score","value":4.3},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":6.3,"unit":"score","value":6.3}],"required_axis_count":3,"score":6.2,"score_formula":"weighted_components_v1","value":15.867,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":1000000},{"key":"gb_second","source":"frontmatter_keyMetrics","unit":"K GB-seconds/mo","value":400000},{"key":"bandwidth_gb","source":"frontmatter_keyMetrics","unit":"GB/mo","value":100}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/digitalocean-functions","slug":"digitalocean-functions","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"DigitalOcean Functions","provider":"DigitalOcean Functions","planName":"freemium","kind":"Serverless Functions","description":"On-demand serverless functions with a flat always-free monthly compute allowance and no idle costs.","logo":"/logos/digitalocean-functions.svg","url":"https://www.digitalocean.com/products/functions","freeLimit":"90,000 GiB-seconds (25 GiB-hours) of compute free every month per account/team, shared across all functions including App Platform function components.","bestFor":"Low-to-moderate-traffic event handlers, webhooks, and API endpoints where you want serverless scale with zero idle cost on the DigitalOcean platform.","caveats":"The free tier is measured purely in compute (GiB-seconds), not invocations, so high-memory or long-running functions exhaust it fast; the allowance does not roll over or prorate, runtime is billed in 100ms increments, and a DigitalOcean account with a payment method on file is required.","sourceUrl":"https://www.digitalocean.com/pricing/functions","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"True","origin":"frontmatter"}],"highlights":["90,000 GiB-seconds (25 GiB-hours) free compute per month, pooled across the whole account/team","True pay-as-you-go beyond the free tier at ~$0.0000185/GiB-s with no reserved instances or idle charges","Free allowance applies regardless of deploy method (console, CLI/doctl, or App Platform components)","No cold-start sleep tier or always-on instance fees, functions run fully on-demand"],"tags":["serverless","functions","faas","digitalocean","compute","pay-as-you-go"],"pricingModel":"freemium","renewal":"monthly","requiresCard":true,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Free tier is 90,000 GiB-seconds (25 GiB-hours) of compute per account/team per month; beyond it, usage is billed at ~$0.0000185/GiB-s. The number of free invocations is derived (e.g. ~3.6M at 256 MiB / 100ms), so high-memory or long-running functions exhaust it faster.","label":"90,000 GiB-seconds/month free compute","renewable":true,"sentiment":"neutral","severity":"medium","type":"total_volume_cap","unit":"GiB-seconds/month","value":90000},{"detail":"The 90,000 GiB-seconds free allowance is not prorated and does not roll over to the next month; unused compute is lost.","label":"Allowance resets monthly, no rollover","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable"},{"detail":"A DigitalOcean account with a credit card or other payment method on file is required; a temporary ~$1 pre-authorization may be placed when adding a card.","label":"Payment method required","renewable":false,"sentiment":"con","severity":"medium","type":"requires_card"},{"detail":"Minimum billed runtime per invocation is 100ms; actual runtime is rounded up to the nearest 100ms.","label":"Runtime billed in 100ms increments","renewable":false,"sentiment":"neutral","severity":"info","type":"performance_throttled","unit":"ms","value":100},{"detail":"Per-function memory is limited to 128 MB – 1 GB, defaulting to 256 MB. Cannot be raised beyond 1 GB.","label":"Memory capped at 1 GB per function","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"MB","value":1024},{"detail":"Maximum timeout per invocation is 15 minutes, including initialization and execution.","label":"15-minute max execution timeout","renewable":false,"sentiment":"con","severity":"low","type":"time_limited","unit":"seconds","value":900},{"detail":"Each namespace is limited to up to 600 invocations per minute and up to 120 concurrent functions.","label":"600 invocations/minute per namespace","renewable":false,"sentiment":"neutral","severity":"medium","type":"rate_limit","unit":"invocations/minute","value":600},{"detail":"You may have up to 10 namespaces, and up to 200 namespace access keys per team/account (including revoked keys).","label":"Up to 10 namespaces per account","renewable":false,"sentiment":"neutral","severity":"info","type":"single_resource_cap","unit":"namespaces","value":10},{"detail":"Max input parameter size is 1 MB, max result response size is 1 MB, max built-function size is 48 MB, and remote build time is capped at 2 minutes.","label":"1 MB payload / 48 MB build size limits","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"MB payload","value":1},{"detail":"Logs are capped at 256 KB per invocation with up to 3 days retention; activation records (async only) are kept 72 hours. PostgreSQL connection pools are not supported, multi-file functions can't be edited in the control panel, and scheduled triggers are in limited private preview (max 3, not on App Platform).","label":"3-day log retention, no PG connection pools","renewable":false,"sentiment":"neutral","severity":"info","type":"feature_restricted"}],"keyMetrics":[{"display":"3.6M/mo","key":"invocations_per_month","label":"Invocations","primary":true,"unit":"M invocations/mo","value":3.6},{"display":"90K GB·s","key":"compute_gb_seconds","label":"Compute","unit":"K GB-seconds/mo","value":90},{"display":"15 min","key":"max_cpu_per_invocation","label":"CPU/call","unit":"ms","value":900000},{"display":"n/a","key":"bandwidth_per_month","label":"Bandwidth","unit":"GB/mo"}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","verifiedAt":"2026-06-17","badges":["🟢 Commercial OK","🔴 One-time","🔴 Card/balance required","🔴 Trial expires","🟠 Single resource"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":2.22,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.7,"unit":"score","value":4.7,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":5.7,"unit":"$/mo","value":2.220003},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":2.7,"unit":"score","value":2.7},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":5.1,"unit":"score","value":5.1}],"required_axis_count":3,"score":4.7,"score_formula":"weighted_components_v1","value":2.22,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":3600000},{"key":"gb_second","source":"frontmatter_keyMetrics","unit":"K GB-seconds/mo","value":90000}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/google-cloud-run-functions","slug":"google-cloud-run-functions","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Google Cloud Run functions","provider":"Google Cloud Run functions","planName":"freemium","kind":"Serverless functions (FaaS)","description":"Event-driven serverless functions on Google Cloud (formerly Cloud Functions) with a perpetual monthly free tier.","logo":"/logos/google-cloud-run-functions.svg","url":"https://cloud.google.com/functions","freeLimit":"Perpetual monthly free tier of 2,000,000 invocations, 400,000 GB-seconds, 200,000 GHz-seconds of compute, and 5 GB of outbound internet data transfer, shared across all functions in the billing account.","bestFor":"Lightweight event-driven backends, webhooks, and HTTP APIs deployed close to Google Cloud services like Pub/Sub, Cloud Storage, and Firestore.","caveats":"A valid billing account (credit card) is required even for free-tier usage, and the free tier does not cover Cloud Build, Container Registry, or Artifact Registry storage incurred when building and storing function images. 1st gen functions have a 540-second (9-minute) max timeout, no request concurrency, and cold starts on scale-up.","sourceUrl":"https://cloud.google.com/functions/pricing-1stgen","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"True","origin":"frontmatter"}],"highlights":["Perpetual (non-expiring) free tier that renews every month, not a one-time trial","2,000,000 invocations plus 400,000 GB-seconds and 200,000 GHz-seconds of compute free per month","Native triggers for Pub/Sub, Cloud Storage, Firestore, and HTTP, with auto-scaling to zero","Free tier is a dollar-equivalent allowance applied account-wide across all functions"],"tags":["serverless","faas","functions","google-cloud","gcp","event-driven","http","webhooks","auto-scaling"],"pricingModel":"freemium","renewal":"monthly","requiresCard":true,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"A valid billing account with a payment method is required even to use the free tier.","label":"Billing account required","renewable":false,"sentiment":"con","severity":"medium","type":"requires_card"},{"detail":"Free tier of 2M invocations, 400,000 GB-seconds, 200,000 GHz-seconds, and 5 GB egress resets each calendar month.","label":"Monthly free allowance","renewable":true,"sentiment":"pro","severity":"info","type":"quota_renewable","unit":"month"},{"detail":"First 2 million invocations per month are free; beyond that they are billed at $0.40 per million.","label":"2M invocations/month","renewable":true,"sentiment":"neutral","severity":"low","type":"total_volume_cap","unit":"invocations/mo","value":2000000},{"detail":"400,000 GB-seconds of memory-time free per month; excess billed at $0.0000025 per GB-second.","label":"400K GB-seconds compute","renewable":true,"sentiment":"neutral","severity":"low","type":"total_volume_cap","unit":"GB-seconds/mo","value":400000},{"detail":"200,000 GHz-seconds of CPU-time free per month; excess billed at $0.0000100 per GHz-second.","label":"200K GHz-seconds compute","renewable":true,"sentiment":"neutral","severity":"low","type":"total_volume_cap","unit":"GHz-seconds/mo","value":200000},{"detail":"5 GB of outbound internet data transfer free per month; excess billed at $0.12 per GB. Inbound and intra-region Google API traffic is free.","label":"5 GB egress/month","renewable":true,"sentiment":"neutral","severity":"low","type":"total_volume_cap","unit":"GB/mo","value":5},{"detail":"Free tier excludes Cloud Build, Container Registry, and Artifact Registry charges for building and storing function images; small storage charges accrue beyond their own free tiers.","label":"Image build/storage not free","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"1st gen functions have a maximum execution timeout of 540 seconds (9 minutes); long-running event-driven functions are terminated at this limit.","label":"540s max timeout","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"seconds","value":540},{"detail":"1st gen functions handle one request per instance with no concurrency support; concurrent traffic requires more instances (use 2nd gen / Cloud Run for concurrency).","label":"No concurrency (1st gen)","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Functions scale to zero when idle, so the first invocation after idle incurs cold-start latency unless minimum instances (billed as idle time) are configured.","label":"Cold starts","renewable":false,"sentiment":"neutral","severity":"low","type":"cold_start_sleep"}],"keyMetrics":[{"display":"2M/mo","key":"invocations_per_month","label":"Invocations","primary":true,"unit":"M invocations/mo","value":2},{"display":"400K GB·s","key":"compute_gb_seconds","label":"Compute","unit":"K GB-seconds/mo","value":400},{"display":"9 min","key":"max_cpu_per_invocation","label":"CPU/call","note":"9 min","unit":"ms","value":540000},{"display":"5 GB/mo","key":"bandwidth_per_month","label":"Bandwidth","unit":"GB/mo","value":5}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","verifiedAt":"2026-06-17","badges":["🟢 Commercial OK","🔴 Card/balance required","🔴 One-time","🟠 Single resource","🟠 Cold start","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":7.52,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":6.2,"unit":"score","value":6.2,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":6.8,"unit":"$/mo","value":7.516680000000001},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":5.3,"unit":"score","value":5.3},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":6.3,"unit":"score","value":6.3}],"required_axis_count":3,"score":6.2,"score_formula":"weighted_components_v1","value":7.517,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":2000000},{"key":"gb_second","source":"frontmatter_keyMetrics","unit":"K GB-seconds/mo","value":400000},{"key":"bandwidth_gb","source":"frontmatter_keyMetrics","unit":"GB/mo","value":5}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/oracle-cloud-functions","slug":"oracle-cloud-functions","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Oracle Cloud Functions","provider":"Oracle Cloud Functions","planName":"freemium","kind":"Serverless FaaS","description":"Serverless functions-as-a-service on Oracle Cloud Infrastructure, built on the open-source Fn Project, with a recurring monthly free allowance.","logo":"/logos/oracle-cloud-functions.png","url":"https://www.oracle.com/cloud/cloud-native/functions/","freeLimit":"2,000,000 function invocations and 400,000 GB memory-seconds of execution time free every month, with no time limit on these monthly amounts.","bestFor":"Event-driven automation, serverless APIs, and intermittent/spiky workloads for developers already on or willing to adopt Oracle Cloud Infrastructure.","caveats":"A valid credit/debit card is required at signup for identity verification (temporary authorization hold, not a charge). Functions is not listed among OCI's \"Always Free only-state\" resources — the free amounts are a recurring monthly Pay-As-You-Go allowance, so usage beyond 2M invocations or 400K GB-seconds is billed automatically. Memory caps at 3 GB and execution time at 300 seconds per invocation. Concurrency is limited by a per-Availability-Domain memory pool (default ~60 GB).","sourceUrl":"https://www.oracle.com/cloud/cloud-native/functions/","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"True","origin":"frontmatter"}],"highlights":["2 million free invocations per month — double AWS Lambda's free 1 million","Recurring monthly free allowance with no expiry on the per-month amounts","Built on the open Apache 2.0 Fn Project, Docker, and CloudEvents — portable, no lock-in","Native support for Python, Go, Java, Node.js, and C#, or bring your own Dockerfile","Pairs with 10 TB/month of free OCI outbound data transfer"],"tags":["serverless","faas","functions","oracle-cloud","oci","fn-project","event-driven","cloud-native"],"pricingModel":"freemium","renewal":"monthly","requiresCard":true,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"A valid credit or debit card (functioning like a credit card) is required to create an Oracle Cloud account for identity verification; a temporary authorization hold may be placed. Prepaid, virtual, or PIN debit cards are not accepted.","label":"Card required at signup","renewable":false,"sentiment":"con","severity":"high","type":"requires_card"},{"detail":"The first 2,000,000 function invocations each month are free; invocations beyond that are billed at US$0.0000002 each. The allowance resets monthly.","label":"2M invocations/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"invocations/mo","value":2000000},{"detail":"The first 400,000 gigabyte memory-seconds of execution time each month are free; usage beyond that is billed at US$0.00001417 per GB-second. The allowance resets monthly.","label":"400K GB-seconds/month","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"GB-seconds/mo","value":400000},{"detail":"Each function can be allocated at most 3 GB (3072 MB) of memory per invocation; selectable values are 128/256/512/1024/2048/3072 MB.","label":"3 GB max memory per function","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"MB","value":3072},{"detail":"A single function invocation can run for at most 300 seconds (5 minutes) before timing out.","label":"300-second timeout","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"seconds","value":300},{"detail":"Concurrency is bounded by a per-Availability-Domain memory pool (default ~60 GB total), so the number of simultaneous executions depends on each function's memory size rather than an unlimited concurrency model.","label":"Per-AD concurrency pool","renewable":false,"sentiment":"neutral","severity":"low","type":"rate_limit","unit":"GB per AD","value":60},{"detail":"OCI Functions is not listed among the resources usable in an Always-Free-only account state; the free invocation and execution-time amounts are a recurring monthly Pay-As-You-Go allowance, and overages are billed automatically once the account is active beyond the trial.","label":"Not in Always-Free-only state","renewable":false,"sentiment":"neutral","severity":"medium","type":"feature_restricted"},{"detail":"Functions that are not kept warm incur cold-start latency; sub-second response during spikes requires paid Provisioned Concurrency (unused provisioned concurrency is billed at 25% of execution-time rates).","label":"Cold starts on idle functions","renewable":false,"sentiment":"neutral","severity":"low","type":"cold_start_sleep"}],"keyMetrics":[{"display":"2M/mo","key":"invocations_per_month","label":"Invocations","primary":true,"unit":"M invocations/mo","value":2},{"display":"400K GB·s","key":"compute_gb_seconds","label":"Compute","unit":"K GB-seconds/mo","value":400},{"display":"5 min","key":"max_cpu_per_invocation","label":"CPU/call","note":"300s timeout","unit":"ms","value":300000},{"display":"10.24 TB/mo","key":"account_egress_gb","label":"Shared egress","note":"OCI account-wide outbound transfer; shown for context, not a Functions-specific allowance.","unit":"GB/mo","value":10240}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","verifiedAt":"2026-06-17","badges":["🟢 Commercial OK","🔴 Card/balance required","🟠 Single resource","🟠 Cold start","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":7.07,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":6.3,"unit":"score","value":6.3,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":6.7,"unit":"$/mo","value":7.066680000000001},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":5.1,"unit":"score","value":5.1},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":7,"unit":"score","value":7}],"required_axis_count":3,"score":6.3,"score_formula":"weighted_components_v1","value":7.067,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":2000000},{"key":"gb_second","source":"frontmatter_keyMetrics","unit":"K GB-seconds/mo","value":400000}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/scaleway-serverless-functions","slug":"scaleway-serverless-functions","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Scaleway Serverless Functions","provider":"Scaleway Serverless Functions","planName":"freemium","kind":"Serverless Functions (FaaS)","description":"European cloud FaaS that runs Go, Node, Python, PHP, and Rust code on demand with a permanent monthly free allowance.","logo":"/logos/scaleway-serverless-functions.png","url":"https://www.scaleway.com/en/serverless-functions/","freeLimit":"Every account gets 1,000,000 requests plus 400,000 GB-seconds of resource consumption free per month, with usage beyond that billed at ~EUR 0.15 per 1M requests and ~EUR 0.50 per 100k GB-s.","bestFor":"Event-driven APIs, webhooks, and lightweight backends that want EU data residency and a real per-month free allowance rather than a one-off trial.","caveats":"The free tier is a usage allowance on a pay-as-you-go account, so a verified payment method is required and any consumption above the monthly allowance is billed automatically. Functions scale to zero after 15 minutes of inactivity, causing cold starts unless you pay for provisioned (warm) instances. There is no documented free egress bandwidth allotment.","sourceUrl":"https://www.scaleway.com/en/pricing/serverless/","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"True","origin":"frontmatter"}],"highlights":["1M requests + 400,000 GB-seconds free every month, per account, with no expiry","Five runtimes supported: Go, Node.js, Python, PHP, and Rust","Scales up to 50 concurrent instances per function and 5,000 invocations/second","EU-based (French) provider with GDPR-friendly data residency"],"tags":["serverless","faas","functions","europe","gdpr","go","nodejs","python","php","rust","pay-as-you-go"],"pricingModel":"freemium","renewal":"monthly","requiresCard":true,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"1,000,000 requests free per account per month; resets monthly. Beyond this, ~EUR 0.15 per 1M requests.","label":"Monthly request allowance","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"requests/mo","value":1000000},{"detail":"400,000 GB-seconds of resource consumption free per account per month; resets monthly. Beyond this, ~EUR 0.000005/GB-s (EUR 0.50 per 100k GB-s).","label":"Monthly compute allowance","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"GB-s/mo","value":400000},{"detail":"The free allowance lives on a pay-as-you-go account; a verified payment method is needed and overage is billed automatically.","label":"Payment method required","renewable":false,"sentiment":"con","severity":"high","type":"requires_card"},{"detail":"Idle function instances scale to zero after 15 minutes (scale down after 30 seconds), incurring cold starts. Keeping instances warm requires paid provisioned resources (~EUR 1.70 per 100k GB-s).","label":"Scale to zero after 15 minutes","renewable":false,"sentiment":"neutral","severity":"medium","type":"cold_start_sleep","unit":"minutes","value":15},{"detail":"Maximum 5,000 invocations per second per function.","label":"Invocation rate cap","renewable":false,"sentiment":"neutral","severity":"low","type":"rate_limit","unit":"req/s","value":5000},{"detail":"Each function instance handles a single request at a time; throughput depends on instances scaling out.","label":"Concurrency of 1 per instance","renewable":false,"sentiment":"neutral","severity":"low","type":"performance_throttled","unit":"request/instance","value":1},{"detail":"A function scales to a maximum of 50 simultaneous instances.","label":"Max 50 instances per function","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"instances","value":50},{"detail":"HTTP request duration ranges from 10 seconds to a maximum of 60 minutes; longer tasks require Serverless Jobs.","label":"Request timeout 60 minutes","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"minutes","value":60},{"detail":"Maximum request payload size is 6 MiB.","label":"Payload size 6 MiB","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"MiB","value":6},{"detail":"Max zip size 100 MiB, max code size 500 MiB per function; 1024 MiB ephemeral temporary disk per instance.","label":"Code/zip size limits","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"MiB","value":500},{"detail":"Outbound traffic on ports 25 and 465 is blocked (except via Scaleway Transactional Email); no dedicated IPs; only HTTP/1.1 and HTTP/2 supported; no versioning/rollback.","label":"Blocked SMTP ports","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"Max 1,000 functions and 600 GiB total function memory per organization, 100 namespaces per project (lower before account verification).","label":"Account-wide function limits","renewable":false,"sentiment":"con","severity":"info","type":"total_volume_cap","unit":"functions","value":1000}],"keyMetrics":[{"display":"1M/mo","key":"invocations_per_month","label":"Invocations","primary":true,"unit":"M invocations/mo","value":1},{"display":"400K GB·s","key":"compute_gb_seconds","label":"Compute","unit":"K GB-seconds/mo","value":400},{"display":"1 hr","key":"max_cpu_per_invocation","label":"CPU/call","note":"3,600,000 ms","unit":"ms","value":3600000},{"display":"n/a","key":"bandwidth_per_month","label":"Bandwidth","unit":"GB/mo"}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","verifiedAt":"2026-06-17","badges":["🟢 Commercial OK","🔴 Card/balance required","🟠 Cold start","🟠 Single resource","🔴 One-time","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":6.87,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":5.6,"unit":"score","value":5.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":6.7,"unit":"$/mo","value":6.866680000000001},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":3.3,"unit":"score","value":3.3},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":6.3,"unit":"score","value":6.3}],"required_axis_count":3,"score":5.6,"score_formula":"weighted_components_v1","value":6.867,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":1000000},{"key":"gb_second","source":"frontmatter_keyMetrics","unit":"K GB-seconds/mo","value":400000}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/vercel-sandbox","slug":"vercel-sandbox","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Vercel Sandbox","provider":"Vercel Sandbox","planName":"freemium","kind":"Ephemeral microVM runner","description":"Isolated, ephemeral Linux microVMs for safely running untrusted or AI-generated code.","logo":"/logos/vercel.svg","url":"https://vercel.com/docs/vercel-sandbox","freeLimit":"Hobby includes 5 Active CPU hours, 5,000 sandbox creations, 20 GB data transfer, and 420 GB-hours of memory per month, plus 15 GB lifetime snapshot storage.","bestFor":"Running AI-agent output, code playgrounds, and untrusted user code in disposable Firecracker microVMs without standing infrastructure.","caveats":"Hobby is personal/noncom only; once any monthly limit is hit, sandbox creation pauses until the cycle resets (30 days from first use). Only the iad1 region is available, and storage is a 15 GB lifetime cap rather than a renewable monthly quota.","sourceUrl":"https://vercel.com/docs/sandbox/pricing","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"False","origin":"frontmatter"}],"highlights":["Firecracker microVM isolation with full root/sudo, Docker, VPN and FUSE support","Millisecond startup with Node and Python runtimes, persistent state on stop/resume","Generous 5,000 free sandbox creations per month on the Hobby tier","JS, Python SDKs and a CLI for programmatic create/run/file workflows"],"tags":["microvm","sandbox","code-execution","ai-agents","untrusted-code","firecracker","serverless"],"pricingModel":"freemium","renewal":"monthly","requiresCard":false,"creditAmountUsd":null,"eligibility":["anyone"],"limitations":[{"detail":"Up to 5,000 Sandbox.create() calls per billing cycle; resets each month.","label":"5,000 creations/mo","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"creations/month","value":5000},{"detail":"5 hours of active CPU time per month (I/O wait is not counted); resets each cycle.","label":"5 Active CPU hrs/mo","renewable":true,"sentiment":"pro","severity":"medium","type":"quota_renewable","unit":"CPU-hours/month","value":5},{"detail":"420 GB-hours of provisioned memory per month; resets each cycle.","label":"420 GB-hrs memory/mo","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"GB-hours/month","value":420},{"detail":"20 GB of inbound/outbound sandbox data transfer per month.","label":"20 GB transfer/mo","renewable":true,"sentiment":"neutral","severity":"low","type":"total_volume_cap","unit":"GB/month","value":20},{"detail":"15 GB snapshot storage as a one-time lifetime cap, not a renewable monthly quota.","label":"15 GB storage (lifetime)","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"GB","value":15},{"detail":"A single sandbox can run for at most 45 minutes on Hobby (Pro/Enterprise allow 5 hours).","label":"45 min max runtime","renewable":false,"sentiment":"con","severity":"medium","type":"time_limited","unit":"minutes","value":45},{"detail":"At most 10 sandboxes can run concurrently on Hobby.","label":"10 concurrent","renewable":false,"sentiment":"neutral","severity":"medium","type":"single_resource_cap","unit":"sandboxes","value":10},{"detail":"Each Hobby sandbox is capped at 4 vCPUs and 8 GB memory (Pro 8/16, Enterprise 32/64).","label":"4 vCPU / 8 GB max","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"vCPU","value":4},{"detail":"vCPU allocation is rate-limited to 40 vCPUs per 10 minutes; control plane capped at 1,000 requests/minute.","label":"40 vCPU/10 min","renewable":true,"sentiment":"neutral","severity":"medium","type":"rate_limit","unit":"vCPU/10min","value":40},{"detail":"Once any included limit is exceeded, sandbox creation is paused until 30 days have passed since first use (cycle reset); upgrade to Pro to continue.","label":"Creation pauses at cap","renewable":true,"sentiment":"neutral","severity":"high","type":"cold_start_sleep"},{"detail":"Vercel Sandbox is currently available only in the iad1 (US East) region.","label":"iad1 region only","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted"},{"detail":"The Hobby plan is intended for personal, noncom use; commercial workloads require Pro or Enterprise.","label":"Personal/noncom","renewable":false,"sentiment":"con","severity":"high","type":"no_commercial_use"}],"keyMetrics":[{"display":"5000M/mo","key":"invocations_per_month","label":"Sandbox creations","primary":true,"unit":"creations/month","value":5000},{"display":"5K GB·s","key":"compute_gb_seconds","label":"Active CPU","unit":"CPU-hours/month","value":5},{"display":"4 ms","key":"max_cpu_per_invocation","label":"Max vCPUs per sandbox","unit":"vCPU","value":4},{"display":"20 GB/mo","key":"bandwidth_per_month","label":"Data transfer","unit":"GB/month","value":20}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","verifiedAt":"2026-06-17","badges":["🔴 Non-commercial","🔴 One-time","🟠 Single resource","🔴 Trial expires","🟠 Cold start","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":1.8,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.4,"unit":"score","value":4.4,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":5.5,"unit":"$/mo","value":1.8010833334999998},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":3.1,"unit":"score","value":3.1},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":3.6,"unit":"score","value":3.6}],"required_axis_count":3,"score":4.4,"score_formula":"weighted_components_v1","value":1.801,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"allowed_use":"noncom","source":"stated","value":false},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"creations/month","value":5000},{"key":"gb_second","source":"frontmatter_keyMetrics","unit":"CPU-hours/month","value":5},{"key":"bandwidth_gb","source":"frontmatter_keyMetrics","unit":"GB/month","value":20}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/yandex-cloud-functions","slug":"yandex-cloud-functions","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Yandex Cloud Functions","provider":"Yandex Cloud Functions","planName":"freemium","kind":"Serverless FaaS","description":"Serverless function-as-a-service that runs your code on demand with no VM management, billed per invocation and GB-hour of execution.","logo":"/logos/yandex-cloud-functions.png","url":"https://yandex.cloud/en/services/functions","freeLimit":"Every month you get 1,000,000 function invocations plus 10 GB-hours of execution time free of charge, with pay-as-you-go pricing after that.","bestFor":"Event-driven backends, webhooks, and low-to-moderate-traffic APIs that fit within a permanent monthly serverless allowance without paying for idle time.","caveats":"Requires a Yandex Cloud billing account (pay-as-you-go beyond the grant); the free allowance resets and any unused balance is forfeited each month; default quotas cap you at 10 functions per cloud and 10 concurrent invocations per availability zone; the platform is primarily oriented to the Russia region.","sourceUrl":"https://yandex.cloud/en/docs/functions/pricing","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"True","origin":"frontmatter"}],"highlights":["Permanent monthly free grant of 1,000,000 invocations + 10 GB-hours, not a one-time trial","Only invocations that actually run your code are billed (no charge for filtered calls)","Triggers (timer, message queue, object storage, etc.) are included free within quotas","Up to 8 GB RAM per instance and execution timeouts up to 1 hour for long-lived functions"],"tags":["serverless","faas","functions","event-driven","yandex-cloud","pay-as-you-go"],"pricingModel":"freemium","renewal":"monthly","requiresCard":true,"creditAmountUsd":null,"eligibility":["anyone","region_restricted"],"limitations":[{"detail":"The 1,000,000 invocations and 10 GB-hours are granted each month; any unused balance is reset at the end of the month and does not roll over.","label":"Monthly free allowance resets","renewable":true,"sentiment":"pro","severity":"info","type":"quota_renewable","unit":"invocations/mo","value":1000000},{"detail":"Free execution time is capped at 10 GB-hours per month (memory in GB multiplied by runtime in hours); beyond this you pay the standard compute rate.","label":"10 GB-hours execution/month","renewable":true,"sentiment":"neutral","severity":"medium","type":"total_volume_cap","unit":"GB-hours/mo","value":10},{"detail":"A Yandex Cloud billing account is required to use the service; usage above the free grant is charged pay-as-you-go.","label":"Billing account required","renewable":false,"sentiment":"con","severity":"medium","type":"requires_card"},{"detail":"Yandex Cloud is primarily oriented to the Russia region; availability, payment methods, and signup may be restricted depending on your location.","label":"Region-oriented (Russia)","renewable":false,"sentiment":"con","severity":"high","type":"requires_eligibility"},{"detail":"Default quota of 10 concurrent function calls and 10 function instances per availability zone (increasable via support request).","label":"10 concurrent invocations per zone","renewable":false,"sentiment":"neutral","severity":"medium","type":"rate_limit","unit":"concurrent calls/zone","value":10},{"detail":"Default quota allows only 10 functions per cloud and 100 triggers per cloud (increasable via support request).","label":"10 functions per cloud","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"functions/cloud","value":10},{"detail":"Maximum 8 GB RAM per function instance, with a default cap of 20 GB total RAM across all running functions per availability zone.","label":"8 GB RAM, 20 GB total per zone","renewable":false,"sentiment":"neutral","severity":"low","type":"single_resource_cap","unit":"GB RAM/instance","value":8},{"detail":"Maximum execution time per call is 1 hour, but timeouts longer than 10 minutes are only available for long-lived functions; request/response JSON is capped at 3.5 MB and temporary files at 512 MB.","label":"1 hour max execution time","renewable":false,"sentiment":"neutral","severity":"low","type":"feature_restricted","unit":"seconds","value":3600},{"detail":"Functions scale to zero when idle and incur cold-start latency on the next call; keeping warm via provisioned instances is billable (and provisioned instances are limited to 2 per cloud by default).","label":"Cold starts on idle","renewable":false,"sentiment":"neutral","severity":"low","type":"cold_start_sleep"}],"keyMetrics":[{"display":"1M/mo","key":"invocations_per_month","label":"Invocations","primary":true,"unit":"M invocations/mo","value":1},{"display":"36K GB·s","key":"compute_gb_seconds","label":"Compute","note":"10 GB-hours","unit":"K GB-seconds/mo","value":36},{"display":"1 hr","key":"max_cpu_per_invocation","label":"CPU/call","unit":"ms","value":3600000},{"display":"n/a","key":"bandwidth_per_month","label":"Bandwidth","note":"billed separately","unit":"GB/mo"}],"verified":true,"timestamp":"2026-06-17T00:00:00+00:00","verifiedAt":"2026-06-17","badges":["🟢 Commercial OK","🔴 One-time","🔴 Card/balance required","🔴 Eligibility required","🟠 Single resource","🟠 Cold start","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"allowance_usd_mo":0.8,"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.4,"unit":"score","value":4.4,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"provider_quantity_times_reference_price","score":4.8,"unit":"$/mo","value":0.8000012000000001},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":3.7,"unit":"score","value":3.7},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":4.5,"unit":"score","value":4.5}],"required_axis_count":3,"score":4.4,"score_formula":"weighted_components_v1","value":0.8,"value_kind":"usd_month","value_provenance":"provider_quantity_times_reference_price"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[{"key":"invocations","source":"frontmatter_keyMetrics","unit":"M invocations/mo","value":1000000},{"key":"gb_second","source":"frontmatter_keyMetrics","unit":"K GB-seconds/mo","value":36000}],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/cloudflare-d1-workers-paid","slug":"cloudflare-d1-workers-paid","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Cloudflare D1","provider":"Cloudflare D1","planName":"paid","kind":"Paid plan","description":"First 25B rows read + 50M rows written + 5GB storage included/month (in $5 Workers Paid); then $0.001/M rows read, $1.00/M rows written, $0.75/GB-month. No egress charges.","logo":"/logos/cloudflare-d1-workers-paid.png","url":"https://developers.cloudflare.com/workers/platform/pricing/","freeLimit":"Paid — $0.001 per 1M rows read; $1.00 per 1M rows written; $0.75/GB-month. First 25B rows read + 50M rows written + 5GB storage included/month (in $5 Workers Paid); then $0.001/M rows read, $1.00/M rows written, $0.75/GB-month. No egress charges.","bestFor":"1 billion rows read per $1 - extraordinarily cheap relational reads; included allowances cover most apps inside the $5 Workers Paid base.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$1/M wr","key":"m0","label":"","primary":true},{"display":"25B rd incl","key":"m1","label":"","primary":false},{"display":"5GB incl","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Best value","🟢 Low/no egress"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/cloudflare-workers-paid-standard","slug":"cloudflare-workers-paid-standard","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Cloudflare Workers","provider":"Cloudflare Workers","planName":"paid","kind":"Paid plan","description":"10M requests + 30M CPU-ms/month included; overage $0.30/M requests, $0.02/M CPU-ms. No duration (I/O wait) charge, no egress, edge in 300+ cities.","logo":"/logos/cloudflare-workers-paid.png","url":"https://developers.cloudflare.com/workers/platform/pricing/","freeLimit":"Paid — $5/month minimum. 10M requests + 30M CPU-ms/month included; overage $0.30/M requests, $0.02/M CPU-ms. No duration (I/O wait) charge, no egress, edge in 300+ cities.","bestFor":"~2M requests per $1 of overage; first 10M requests effectively free inside the $5 base. Among the cheapest per-request edge compute available.","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":5,"valueText":"5.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$5/mo min","key":"m0","label":"","primary":true},{"display":"10M req","key":"m1","label":"","primary":false},{"display":"30M CPU-ms","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Best value","🟢 Low/no egress","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":5,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/saladcloud","slug":"saladcloud","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"SaladCloud","provider":"SaladCloud","planName":"paid","kind":"Paid plan","description":"Distributed consumer-GPU cloud billed per second; RTX 4090 $0.16/hr, RTX 5090 $0.25/hr, H100 NVL $0.99/hr. 10,000+ edge GPUs.","logo":"/logos/saladcloud.png","url":"https://salad.com/","freeLimit":"Paid — From $0.10/hr; RTX 4090 $0.16/hr; H100 NVL $0.99/hr; RTX 5090 $0.25/hr. Distributed consumer-GPU cloud billed per second; RTX 4090 $0.16/hr, RTX 5090 $0.25/hr, H100 NVL $0.99/hr. 10,000+ edge GPUs.","bestFor":"The cheapest GPU rates on the market - H100 NVL at $0.99/hr and RTX 4090 at $0.16/hr - via distributed consumer hardware. Extraordinary raw value for ","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"from $.10/h","key":"m0","label":"","primary":true},{"display":"4090 $.16/h","key":"m1","label":"","primary":false},{"display":"H100 $.99","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Best value"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/vast-ai","slug":"vast-ai","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Vast.ai","provider":"Vast.ai","planName":"paid","kind":"Paid plan","description":"Marketplace of 20,000+ GPUs with supply/demand pricing; RTX 4090 from $0.43/hr (interruptible ~$0.31/hr), H200 from $3.55/hr, B200 from $4.25/hr; any Docker image.","logo":"/logos/vast-ai.png","url":"https://vast.ai/pricing","freeLimit":"Paid — RTX 4090 from $0.43/hr; H100 SXM from ~$1.65/hr; RTX 5090 from $0.66/hr. Marketplace of 20,000+ GPUs with supply/demand pricing; RTX 4090 from $0.43/hr (interruptible ~$0.31/hr), H200 from $3.55/hr, B200 from $4.25/hr; any Docker image.","bestFor":"Among the absolute cheapest GPU compute anywhere - RTX 4090 from ~$0.31-0.43/hr and H100s from ~$1.65/hr; unbeatable raw value if you accept variable ","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"4090 $.43/h","key":"m0","label":"","primary":true},{"display":"H100 $1.65","key":"m1","label":"","primary":false},{"display":"5090 $.66","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Best value"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/cloudflare-durable-objects-workers-paid","slug":"cloudflare-durable-objects-workers-paid","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Cloudflare Durable Objects","provider":"Cloudflare Durable Objects","planName":"paid","kind":"Paid plan","description":"1M requests + 400,000 GB-s/month included; overage $0.15/M requests, $12.50/M GB-s. Stateful single-threaded objects with hibernation (no duration billing while hibernating).","logo":"/logos/cloudflare-durable-objects-workers-paid.png","url":"https://developers.cloudflare.com/workers/platform/pricing/","freeLimit":"Paid — $0.15 per 1M requests + $12.50 per 1M GB-s (in $5 Workers Paid). 1M requests + 400,000 GB-s/month included; overage $0.15/M requests, $12.50/M GB-s. Stateful single-threaded objects with hibernation (no duration billing while hibernating).","bestFor":"~6.7M requests per $1 of overage; hibernation means idle stateful agents cost almost nothing. Excellent for long-lived per-user/agent state.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$0.15/M req","key":"m0","label":"","primary":true},{"display":"1M req incl","key":"m1","label":"","primary":false},{"display":"400k GB-s","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/cloudflare-r2-workers-paid","slug":"cloudflare-r2-workers-paid","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Cloudflare Containers GPU note / R2","provider":"Cloudflare Containers GPU note / R2","planName":"paid","kind":"Paid plan","description":"$5/mo unlocks Workers + Containers + Durable Objects + R2 ($0.015/GB-month standard, Class A $4.50/M, Class B $0.36/M) with no egress fees.","logo":"/logos/cloudflare-r2-workers-paid.png","url":"https://www.cloudflare.com/plans/developer-platform/","freeLimit":"Paid — $5/month base; R2 storage $0.015/GB-month, zero egress. $5/mo unlocks Workers + Containers + Durable Objects + R2 ($0.015/GB-month standard, Class A $4.50/M, Class B $0.36/M) with no egress fees.","bestFor":"Zero egress is a major saving vs hyperscalers (S3 egress ~$0.09/GB); $0.015/GB-month storage plus free bandwidth is exceptional value for data-heavy s","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":5,"valueText":"5.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$5/mo","key":"m0","label":"","primary":true},{"display":"R2 $.015/GB","key":"m1","label":"","primary":false},{"display":"$0 egress","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Low/no egress","🟢 Embeddings/RAG"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":5,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/cloudflare-workers-kv-paid","slug":"cloudflare-workers-kv-paid","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Cloudflare Workers KV","provider":"Cloudflare Workers KV","planName":"paid","kind":"Paid plan","description":"10M reads + 1M writes + 1GB storage included/month; reads $0.50/M, writes/deletes/lists $5/M, storage $0.50/GB-month. Low-latency global reads.","logo":"/logos/cloudflare-workers-kv-paid.png","url":"https://developers.cloudflare.com/workers/platform/pricing/","freeLimit":"Paid — $0.50 per 1M reads; $5 per 1M writes; $0.50/GB-month storage. 10M reads + 1M writes + 1GB storage included/month; reads $0.50/M, writes/deletes/lists $5/M, storage $0.50/GB-month. Low-latency global reads.","bestFor":"~2M reads per $1; included allowances cover most small apps inside the $5 Workers Paid base. Cheap durable state for edge functions/agents.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$.50/M rd","key":"m0","label":"","primary":true},{"display":"$5/M wr","key":"m1","label":"","primary":false},{"display":"10M rd incl","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/cloudflare-queues-workers-paid","slug":"cloudflare-queues-workers-paid","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Cloudflare Queues","provider":"Cloudflare Queues","planName":"paid","kind":"Paid plan","description":"1M operations/month included, then $0.40/M operations (one op per 64KB written/read/deleted). No egress/bandwidth charges.","logo":"/logos/cloudflare-queues-workers-paid.png","url":"https://developers.cloudflare.com/workers/platform/pricing/","freeLimit":"Paid — $0.40 per 1M operations. 1M operations/month included, then $0.40/M operations (one op per 64KB written/read/deleted). No egress/bandwidth charges.","bestFor":"~2.5M operations per $1 (~833k messages at 3 ops each); included 1M ops in the $5 Workers Paid base. Cheap glue for event-driven serverless code.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$0.40/M op","key":"m0","label":"","primary":true},{"display":"1M op incl","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Low/no egress"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/fly-io-machines","slug":"fly-io-machines","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Fly.io Machines","provider":"Fly.io Machines","planName":"paid","kind":"Paid plan","description":"Per-second billing for VM presets: shared-cpu-1x/256MB ~$0.0000008/s (~$1.94/mo always-on); performance-1x/2GB ~$32/mo. Stopped Machines billed only for storage; no plan fee.","logo":"/logos/fly-io-machines-performance-1x.png","url":"https://fly.io/docs/about/pricing/","freeLimit":"Paid — From ~$1.94/month (shared-cpu-1x 256MB) + ~$5/GB extra RAM. Per-second billing for VM presets: shared-cpu-1x/256MB ~$0.0000008/s (~$1.94/mo always-on); performance-1x/2GB ~$32/mo. Stopped Machines billed only for storage; no plan fee.","bestFor":"An always-on micro-VM for under $2/month; per-second billing means scale-to-zero workloads pay near-nothing. Excellent value for small persistent agen","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":1.94,"valueText":"1.94","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"~$1.94/mo","key":"m0","label":"","primary":true},{"display":"+$5/GB RAM","key":"m1","label":"","primary":false},{"display":"per-sec","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Flat rate"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":1.94,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/runpod-pods-secure-community-cloud","slug":"runpod-pods-secure-community-cloud","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"RunPod Pods","provider":"RunPod Pods","planName":"paid","kind":"Paid plan","description":"On-demand dedicated GPU pods billed per second: A100 PCIe 80GB from $1.39/hr, A40 48GB ~$0.39/hr, L4 24GB $0.39/hr; community cloud cheaper than secure cloud. Any container, persis","logo":"/logos/runpod-pods-secure-community-cloud.png","url":"https://www.runpod.io/pricing","freeLimit":"Paid — A100 PCIe from $1.39/hr; A40 from $0.39/hr; L4 $0.39/hr. On-demand dedicated GPU pods billed per second: A100 PCIe 80GB from $1.39/hr, A40 48GB ~$0.39/hr, L4 24GB $0.39/hr; community cloud cheaper than secure cloud. Any container, persistent storage.","bestFor":"A100 PCIe from $1.39/hr and 48GB A40 at $0.39/hr deliver excellent dedicated-GPU value with per-second billing; community tier among the cheapest reli","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"A100 $1.39/h","key":"m0","label":"","primary":true},{"display":"A40 $0.39/h","key":"m1","label":"","primary":false},{"display":"L4 $0.39/h","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/koyeb-serverless-gpu-single-gpu","slug":"koyeb-serverless-gpu-single-gpu","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Koyeb Serverless GPU","provider":"Koyeb Serverless GPU","planName":"paid","kind":"Paid plan","description":"Per-second serverless GPU with scale-to-zero: RTX-4000-SFF-ADA $0.50/hr (6 vCPU/44GB), L4 $0.70/hr (6 vCPU/32GB), RTX-A6000 48GB $0.75/hr (6 vCPU/64GB). Autoscaling endpoints.","logo":"/logos/koyeb-serverless-gpu-single-gpu.png","url":"https://www.koyeb.com/pricing","freeLimit":"Paid — L4 $0.70/hr; RTX-A6000 $0.75/hr; RTX-4000-SFF-ADA $0.50/hr. Per-second serverless GPU with scale-to-zero: RTX-4000-SFF-ADA $0.50/hr (6 vCPU/44GB), L4 $0.70/hr (6 vCPU/32GB), RTX-A6000 48GB $0.75/hr (6 vCPU/64GB). Autoscaling endpoints.","bestFor":"L4 at $0.70/hr and 48GB RTX-A6000 at $0.75/hr with true scale-to-zero are among the best low-end serverless GPU deals; per-second billing for bursty i","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"L4 $0.70/hr","key":"m0","label":"","primary":true},{"display":"A6000 $.75","key":"m1","label":"","primary":false},{"display":"4000 $.50","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/google-cloud-run-gpu-nvidia-l4","slug":"google-cloud-run-gpu-nvidia-l4","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Google Cloud Run GPU","provider":"Google Cloud Run GPU","planName":"paid","kind":"Paid plan","description":"Serverless L4 GPU on Cloud Run: ~$0.0001867/s (~$0.67/hr) without zonal redundancy in Tier 1 regions, plus vCPU/memory; scale-to-zero, no quota request needed.","logo":"/logos/google-cloud-run-gpu-nvidia-l4.png","url":"https://cloud.google.com/run/pricing","freeLimit":"Paid — ~$0.0001867 per GPU-second (NVIDIA L4, Tier 1). Serverless L4 GPU on Cloud Run: ~$0.0001867/s (~$0.67/hr) without zonal redundancy in Tier 1 regions, plus vCPU/memory; scale-to-zero, no quota request needed.","bestFor":"L4 GPU at ~$0.67/hr with true scale-to-zero serverless is excellent value for intermittent inference; you pay only while requests run (if min-instance","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"L4 ~$.67/hr","key":"m0","label":"","primary":true},{"display":"$187n/GPU-s","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/northflank","slug":"northflank","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Northflank","provider":"Northflank","planName":"paid","kind":"Paid plan","description":"Pay-per-use: ~$0.01667/vCPU/hr, ~$0.00833/GB-mem/hr, per-second billing; NVMe $0.15/GB-month. On-demand GPUs: A100 40GB $1.42/hr, A100 80GB $1.76/hr, H100 $2.74/hr, H200 $3.14/hr.","logo":"/logos/northflank.png","url":"https://northflank.com/pricing","freeLimit":"Paid — $0.01667 per vCPU-hour + $0.00833 per GB-RAM-hour. Pay-per-use: ~$0.01667/vCPU/hr, ~$0.00833/GB-mem/hr, per-second billing; NVMe $0.15/GB-month. On-demand GPUs: A100 40GB $1.42/hr, A100 80GB $1.76/hr, H100 $2.74/hr, H200 $3.14/hr.","bestFor":"~60 vCPU-hours per $1 of compute; A100 40GB at $1.42/hr is one of the lowest managed-platform A100 rates. Strong CPU and GPU value.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$.017/vCPUh","key":"m0","label":"","primary":true},{"display":"$.008/GBh","key":"m1","label":"","primary":false},{"display":"per-sec","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟠 Managed option","🟠 BYOK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/runpod-serverless-flex","slug":"runpod-serverless-flex","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"RunPod Serverless","provider":"RunPod Serverless","planName":"paid","kind":"Paid plan","description":"Pay-per-second worker start to stop: RTX 4000/A4000 16GB $0.00016/s; A100 80GB $0.00076/s (~$2.74/hr); H100 PRO 80GB $0.00116/s (~$4.18/hr); B200 $0.00240/s. Scale-to-zero flex wor","logo":"/logos/runpod-serverless-flex.png","url":"https://docs.runpod.io/serverless/pricing","freeLimit":"Paid — From $0.00016/sec (16GB GPU); A100 80GB $0.00076/sec, H100 PRO $0.00116/sec. Pay-per-second worker start to stop: RTX 4000/A4000 16GB $0.00016/s; A100 80GB $0.00076/s (~$2.74/hr); H100 PRO 80GB $0.00116/s (~$4.18/hr); B200 $0.00240/s. Scale-to-zero flex workers.","bestFor":"A100 80GB at ~$2.74/hr flex (active workers ~$2.17/hr with discounts); claims ~25% cheaper than other serverless GPU clouds on flex workers. Excellent","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"16GB $160u/s","key":"m0","label":"","primary":true},{"display":"A100 $760u/s","key":"m1","label":"","primary":false},{"display":"H100 $1.16m","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/cerebrium","slug":"cerebrium","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Cerebrium","provider":"Cerebrium","planName":"paid","kind":"Paid plan","description":"Pay-per-second GPU/CPU: T4 $0.000164/s, L4 $0.000222/s, A10 $0.000306/s, A100 80GB $0.000583/s (~$2.10/hr), H100 $0.000944/s (~$3.40/hr), B200 $0.00167/s; CPU $0.00000655/vCPU/s.","logo":"/logos/cerebrium.png","url":"https://cerebrium.ai/pricing","freeLimit":"Paid — A100 80GB $0.000583/s; H100 $0.000944/s; T4 $0.000164/s. Pay-per-second GPU/CPU: T4 $0.000164/s, L4 $0.000222/s, A10 $0.000306/s, A100 80GB $0.000583/s (~$2.10/hr), H100 $0.000944/s (~$3.40/hr), B200 $0.00167/s; CPU $0.00000655/vCPU/s.","bestFor":"A100 80GB at ~$2.10/hr is one of the lowest serverless A100 rates; 1-3s autoscaling and snapshotting minimize billable idle. Excellent value for burst","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"A100 $.58k/s","key":"m0","label":"","primary":true},{"display":"T4 $164u/s","key":"m1","label":"","primary":false},{"display":"per-sec","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/cloudflare-containers-workers-paid","slug":"cloudflare-containers-workers-paid","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Cloudflare Containers","provider":"Cloudflare Containers","planName":"paid","kind":"Paid plan","description":"Billed per 10ms active: 25 GiB-hours memory + 375 vCPU-minutes + 200 GB-hours disk included/month in the $5 Workers Paid plan; overage at listed rates. Scale-to-zero.","logo":"/logos/cloudflare-containers-workers-paid.png","url":"https://developers.cloudflare.com/workers/platform/pricing/","freeLimit":"Paid — $0.0000025/GiB-s memory, $0.000020/vCPU-s, $0.00000007/GB-s disk (in $5 Workers Paid). Billed per 10ms active: 25 GiB-hours memory + 375 vCPU-minutes + 200 GB-hours disk included/month in the $5 Workers Paid plan; overage at listed rates. Scale-to-zero.","bestFor":"~50,000 vCPU-seconds (~13.9 vCPU-hours) per $1 of overage; integrated with the Workers edge and no extra plan fee beyond the $5 Workers Paid base.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$20u/vCPU-s","key":"m0","label":"","primary":true},{"display":"$2.5u/GiB-s","key":"m1","label":"","primary":false},{"display":"25 GiB-h","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/beam-cloud","slug":"beam-cloud","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Beam Cloud","provider":"Beam Cloud","planName":"paid","kind":"Paid plan","description":"Pay-per-second (only while container runs): CPU $0.19/core/hr, RAM $0.0000056/GB/s, RTX 4090 $0.69/hr, A10G $1.05/hr, H100 $3.50/hr. Scale-to-zero; no charge for image pull or boot","logo":"/logos/beam-cloud.png","url":"https://docs.beam.cloud/v2/resources/pricing-and-billing","freeLimit":"Paid — RTX 4090 $0.000192/s; A10G $0.000292/s; H100 $0.000972/s. Pay-per-second (only while container runs): CPU $0.19/core/hr, RAM $0.0000056/GB/s, RTX 4090 $0.69/hr, A10G $1.05/hr, H100 $3.50/hr. Scale-to-zero; no charge for image pull or boot wait.","bestFor":"RTX 4090 at $0.69/hr and H100 at $3.50/hr with fast cold starts; not billing for image pull/boot is a real saving. Strong value for Python GPU functio","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"4090 $192u/s","key":"m0","label":"","primary":true},{"display":"H100 $972u/s","key":"m1","label":"","primary":false},{"display":"per-sec","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/fal-ai-compute-serverless","slug":"fal-ai-compute-serverless","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"fal.ai","provider":"fal.ai","planName":"paid","kind":"Paid plan","description":"Custom deployments on a GPU fleet from $1.89/hr for H100 (16 vCPU, 200GB RAM, 1TB SSD); Serverless model APIs billed per output (~$0.025/image, $0.05-$0.40/video-sec) or per-second","logo":"/logos/apis/fal-ai.png","url":"https://fal.ai/pricing","freeLimit":"Paid — From $1.89/hr for H100 (Compute); per-output for Serverless. Custom deployments on a GPU fleet from $1.89/hr for H100 (16 vCPU, 200GB RAM, 1TB SSD); Serverless model APIs billed per output (~$0.025/image, $0.05-$0.40/video-sec) or per-second fallback.","bestFor":"H100 from $1.89/hr is among the cheapest H100 rates on a managed inference platform; per-output pricing gives exact cost-per-result for media gen.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"H100 $1.89/hr","key":"m0","label":"","primary":true},{"display":"per-output","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟠 Managed option"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/scaleway-serverless-functions-containers","slug":"scaleway-serverless-functions-containers","category":"serverless-functions","access":"conditional","accessBucket":"free","name":"Scaleway Serverless","provider":"Scaleway Serverless","planName":"freemium","kind":"Paid plan","description":"EUR 0.50 per 100k GB-s after a 400,000 GB-s/month free tier per account; provisioned EUR 0.000017/GB-s. EU-based, GDPR-friendly.","logo":"/logos/scaleway-serverless-functions-containers.png","url":"https://www.scaleway.com/en/pricing/serverless/","freeLimit":"Free 400,000 GB-s/month per account; then EUR 0.50 per 100k GB-s (EUR 0.000005/GB-s, ~$0.0000054), provisioned EUR 0.000017/GB-s. EU-based, GDPR-friendly.","bestFor":"~185,000 GB-s per $1 for on-demand consumption. Large 400k GB-s free tier makes it extremely cheap for European workloads.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"freemium","keyMetrics":[{"display":"$5.4n/GB-s","key":"m0","label":"","primary":true},{"display":"400k GB-s","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.7,"unit":"score","value":4.7,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$/mo","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8.8,"unit":"score","value":8.8}],"required_axis_count":3,"score":4.7,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"usd_month","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/aws-lambda-graviton2-arm","slug":"aws-lambda-graviton2-arm","category":"serverless-functions","access":"conditional","accessBucket":"free","name":"AWS Lambda","provider":"AWS Lambda","planName":"freemium","kind":"Paid plan","description":"Lambda on Graviton2 (Arm) for up to ~34% better price/perf vs x86; duration ~$0.0000133/GB-s, same $0.20/M requests, shares the 1M requests + 400,000 GB-s free tier.","logo":"/logos/aws-lambda.png","url":"https://aws.amazon.com/lambda/pricing/","freeLimit":"Free tier of 1M requests + 400,000 GB-seconds/month (shared); then $0.0000133334 per GB-second (Arm) + $0.20 per 1M requests. Lambda on Graviton2 (Arm) for up to ~34% better price/perf vs x86.","bestFor":"~75,000 GB-seconds (~20.8 GB-hours at 1GB) per $1 of duration on Arm - meaningfully cheaper than x86 for the same memory; same request value.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"freemium","keyMetrics":[{"display":"$13n/GB-s","key":"m0","label":"","primary":true},{"display":"$0.20/M req","key":"m1","label":"","primary":false},{"display":"Arm -34%","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟠 Usage-metered"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.7,"unit":"score","value":4.7,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$/mo","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8.8,"unit":"score","value":8.8}],"required_axis_count":3,"score":4.7,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"usd_month","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/paid-aws-lambda","slug":"paid-aws-lambda","category":"serverless-functions","access":"conditional","accessBucket":"free","name":"AWS Lambda","provider":"AWS Lambda","planName":"paid","kind":"Paid plan","description":"Pay-per-use: $0.20/M requests plus duration in 1ms increments per GB-second. Free tier 1M requests + 400,000 GB-seconds/month. Arm/Graviton2 up to ~20% cheaper.","logo":"/logos/aws-lambda.png","url":"https://aws.amazon.com/lambda/pricing/","freeLimit":"Free tier of 1M requests + 400,000 GB-seconds/month; then $0.20 per 1M requests + ~$0.0000166667 per GB-second (x86, per-use in 1ms increments). Arm/Graviton2 up to ~20% cheaper.","bestFor":"~5M requests per $1 (requests alone); ~60,000 GB-seconds (~16.7 hours at 1GB) per $1 of duration. Deep free tier and ecosystem integration raise effec","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"freemium","keyMetrics":[{"display":"$0.20/M req","key":"m0","label":"","primary":true},{"display":"$16.7n/GB-s","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟠 Usage-metered"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.7,"unit":"score","value":4.7,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$/mo","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8.8,"unit":"score","value":8.8}],"required_axis_count":3,"score":4.7,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"usd_month","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/spheron-network","slug":"spheron-network","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Spheron Network","provider":"Spheron Network","planName":"paid","kind":"Paid plan","description":"Decentralized GPU compute on-demand: H100 PCIe $2.01/hr, RTX 4090 $0.55/hr, A100 effective ~$2.01/hr, B200 from $2.12/hr; pay-per-use, deploy containers.","logo":"/logos/spheron-network.png","url":"https://www.spheron.network/","freeLimit":"Paid — H100 PCIe $2.01/hr on-demand; RTX 4090 $0.55/hr; B200 from $2.12/hr. Decentralized GPU compute on-demand: H100 PCIe $2.01/hr, RTX 4090 $0.55/hr, A100 effective ~$2.01/hr, B200 from $2.12/hr; pay-per-use, deploy containers.","bestFor":"H100 PCIe at $2.01/hr and B200 from $2.12/hr undercut most managed GPU clouds; strong value for on-demand GPU with simple deploys.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"H100 $2.01/h","key":"m0","label":"","primary":true},{"display":"4090 $.55/h","key":"m1","label":"","primary":false},{"display":"B200 $2.12","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/gmi-cloud","slug":"gmi-cloud","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"GMI Cloud","provider":"GMI Cloud","planName":"paid","kind":"Paid plan","description":"On-demand and serverless GPU inference: H100 at $2.10/hr and next-gen H200 at $2.60/hr (container); instant serverless inference deployment, 5-15 min provisioning.","logo":"/logos/gmi-cloud.png","url":"https://www.gmicloud.ai/","freeLimit":"Paid — H100 $2.10/hr; H200 $2.60/hr (container). On-demand and serverless GPU inference: H100 at $2.10/hr and next-gen H200 at $2.60/hr (container); instant serverless inference deployment, 5-15 min provisioning.","bestFor":"H100 at $2.10/hr and H200 at $2.60/hr are well below most managed providers; strong value for modern GPU inference with quick spin-up.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"H100 $2.10/h","key":"m0","label":"","primary":true},{"display":"H200 $2.60","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/google-cloud-run","slug":"google-cloud-run","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Google Cloud Run","provider":"Google Cloud Run","planName":"paid","kind":"Paid plan","description":"Pay-per-use container compute by vCPU-s and GiB-s plus $0.40/M requests. Free tier: 180,000 vCPU-s, 360,000 GiB-s, 2M requests/month. Scale-to-zero; request- or instance-based bill","logo":"/logos/google-cloud-run.png","url":"https://cloud.google.com/run/pricing","freeLimit":"Paid — $0.000024/vCPU-second, $0.0000025/GiB-second, $0.40 per 1M requests. Pay-per-use container compute by vCPU-s and GiB-s plus $0.40/M requests. Free tier: 180,000 vCPU-s, 360,000 GiB-s, 2M requests/month. Scale-to-zero; request- or instance-based billing.","bestFor":"~2.5M requests per $1; ~41,000 vCPU-seconds (~11.5 vCPU-hours) per $1. Generous free tier and full container support add real value.","caveats":"","sourceUrl":"https://cloud.google.com/run/pricing","offerFacts":[{"key":"offered_price_usd","valueNumber":0.000024,"valueText":"2.4e-05","origin":"value_model_raw"},{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"True","origin":"frontmatter"}],"tags":["paid","value","serverless-functions"],"pricingModel":"freemium","renewal":"monthly","requiresCard":true,"eligibility":["anyone"],"limitations":[{"detail":"The free tier includes 180,000 vCPU-seconds, 360,000 GiB-seconds and 2M requests per month; overage is per-resource-second plus $0.40/M requests.","label":"180k vCPU-s + 2M requests/mo free","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable","unit":"requests/mo","value":2000000},{"detail":"Requires a Google Cloud billing account with a card on file.","label":"GCP billing account required","renewable":false,"sentiment":"con","severity":"medium","type":"requires_card"}],"keyMetrics":[{"display":"$24u/vCPU-s","key":"m0","label":"","primary":true},{"display":"$2.5u/GiB-s","key":"m1","label":"","primary":false},{"display":"$0.40/M req","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$/mo","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.8,"unit":"score","value":8.8},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":7.5,"unit":"score","value":7.5}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"usd_month","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":0.000024,"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/paid-modal","slug":"paid-modal","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Modal","provider":"Modal","planName":"paid","kind":"Paid plan","description":"Pay-per-second (no idle billing): T4 $0.000164/s, A100 80GB $0.000694/s (~$2.50/hr), H100 $0.001097/s (~$3.95/hr), B200 $0.001736/s; CPU $0.0000131/core/s. Starter $0 + $30/mo free","logo":"/logos/modal.png","url":"https://modal.com/pricing","freeLimit":"Paid — H100 $0.001097/s; A100 80GB $0.000694/s; T4 $0.000164/s. Pay-per-second (no idle billing): T4 $0.000164/s, A100 80GB $0.000694/s (~$2.50/hr), H100 $0.001097/s (~$3.95/hr), B200 $0.001736/s; CPU $0.0000131/core/s. Starter $0 + $30/mo free credits.","bestFor":"A100 80GB ~$2.50/hr, H100 ~$3.95/hr; $30/mo free credits on the free Starter plan sweeten entry. Premium DX justifies a modest rate premium over bare ","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":0,"valueText":"0.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"H100 $1.1m/s","key":"m0","label":"","primary":true},{"display":"A100 ~$2.50","key":"m1","label":"","primary":false},{"display":"T4 $164u/s","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":0,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/azure-functions-flex-consumption","slug":"azure-functions-flex-consumption","category":"serverless-functions","access":"conditional","accessBucket":"conditional","tier":"limited","name":"Azure Functions","provider":"Azure Functions","planName":"paid","kind":"Paid plan","description":"Flex Consumption: $0.000004/GB-s idle baseline and $0.000016/GB-s execution, plus $0.40/M executions. Free grant of 250,000 executions and 100,000 GB-s/month per subscription.","logo":"/logos/azure-functions-flex-consumption.png","url":"https://azure.microsoft.com/en-us/pricing/details/functions/","freeLimit":"Paid — $0.000016 per GB-s execution + $0.40 per 1M executions. Flex Consumption: $0.000004/GB-s idle baseline and $0.000016/GB-s execution, plus $0.40/M executions. Free grant of 250,000 executions and 100,000 GB-s/month per subscription.","bestFor":"~2.5M executions per $1; ~62,500 GB-seconds per $1 of execution time. Free grant covers light workloads entirely.","caveats":"","sourceUrl":"https://azure.microsoft.com/en-us/pricing/details/functions/","offerFacts":[{"key":"offered_price_usd","valueNumber":0.000016,"valueText":"1.6e-05","origin":"value_model_raw"},{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"},{"key":"renewal","valueText":"monthly","origin":"frontmatter"},{"key":"requiresCard","valueText":"True","origin":"frontmatter"}],"tags":["paid","value","serverless-functions"],"pricingModel":"freemium","renewal":"monthly","requiresCard":true,"eligibility":["anyone"],"limitations":[{"detail":"Flex Consumption includes a monthly free grant of executions and GB-seconds; overage is $0.000016/GB-s execution plus $0.40 per 1M executions.","label":"Free grant of executions/GB-s","renewable":true,"sentiment":"pro","severity":"low","type":"quota_renewable"},{"detail":"Requires an Azure subscription with a payment method on file.","label":"Azure billing account required","renewable":false,"sentiment":"con","severity":"medium","type":"requires_card"},{"detail":"Flex Consumption scales to zero and cold-starts on the next request unless an idle baseline is provisioned.","label":"Cold starts on scale-from-zero","renewable":false,"sentiment":"neutral","severity":"low","type":"cold_start_sleep"}],"keyMetrics":[{"display":"$16n/GB-s","key":"m0","label":"","primary":true},{"display":"$0.40/M ex","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🟢 Commercial OK","🟠 Usage-metered"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4,"unit":"score","value":4,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$/mo","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":6.6,"unit":"score","value":6.6},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":7.5,"unit":"score","value":7.5}],"required_axis_count":3,"score":4,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"usd_month","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":0.000016,"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/paid-digitalocean-functions","slug":"paid-digitalocean-functions","category":"serverless-functions","access":"conditional","accessBucket":"free","name":"DigitalOcean Functions","provider":"DigitalOcean Functions","planName":"paid","kind":"Paid plan","description":"Pay-per-use: $0.0000185/GiB-second of memory+runtime. Free allowance of 90,000 GiB-seconds (25 GiB-hours)/month per team across all functions. No per-request fee.","logo":"/logos/digitalocean-functions.svg","url":"https://www.digitalocean.com/pricing/functions","freeLimit":"Free allowance of 90,000 GiB-seconds (25 GiB-hours)/month per team across all functions, no per-request fee; then pay-per-use $0.0000185/GiB-second of memory+runtime.","bestFor":"~54,000 GiB-seconds (~15 GiB-hours) per $1. Simple flat rate with no per-request fee makes light workloads very cheap; large free allowance.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"freemium","keyMetrics":[{"display":"$18.5n/GB-s","key":"m0","label":"","primary":true},{"display":"90k GiB-s","key":"m1","label":"","primary":false},{"display":"25 GiB-h","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.7,"unit":"score","value":4.7,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$/mo","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8.8,"unit":"score","value":8.8}],"required_axis_count":3,"score":4.7,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"usd_month","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/modal-sandboxes-notebooks","slug":"modal-sandboxes-notebooks","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Modal Sandboxes / Notebooks","provider":"Modal Sandboxes / Notebooks","planName":"paid","kind":"Paid plan","description":"On-demand sandboxed compute for arbitrary/agent code: CPU $0.00003942/core/s, memory $0.00000672/GB/s, GPUs at standard per-second rates; burst without pre-allocating. Pay only for","logo":"/logos/modal-sandboxes-notebooks.png","url":"https://modal.com/pricing","freeLimit":"Paid — CPU $0.00003942/core/s; Memory $0.00000672/GB/s (+ GPU at standard rates). On-demand sandboxed compute for arbitrary/agent code: CPU $0.00003942/core/s, memory $0.00000672/GB/s, GPUs at standard per-second rates; burst without pre-allocating. Pay only for use.","bestFor":"Sub-second-billed sandboxes mean an agent spinning up code execution pays cents; no idle charges. Great value for ephemeral code-execution/agent tool ","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$39u/core-s","key":"m0","label":"","primary":true},{"display":"$6.7u/GB-s","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/lambda-labs-on-demand-cloud","slug":"lambda-labs-on-demand-cloud","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Lambda On-Demand Cloud","provider":"Lambda On-Demand Cloud","planName":"paid","kind":"Paid plan","description":"On-demand hourly GPU instances: V100 16GB $0.79/hr, A100 40GB $1.99/hr, A100 SXM 80GB $2.79/hr, H100 SXM $3.99/hr, B200 SXM6 $6.69/hr. 1-Click Clusters $5.54-$6.16/H100/hr.","logo":"/logos/lambda-labs-on-demand-cloud.png","url":"https://lambda.ai/pricing","freeLimit":"Paid — H100 SXM $3.99/GPU/hr; A100 80GB $2.79/hr; A100 40GB $1.99/hr. On-demand hourly GPU instances: V100 16GB $0.79/hr, A100 40GB $1.99/hr, A100 SXM 80GB $2.79/hr, H100 SXM $3.99/hr, B200 SXM6 $6.69/hr. 1-Click Clusters $5.54-$6.16/H100/hr.","bestFor":"A100 80GB at $2.79/hr and H100 at $3.99/hr are low for dedicated on-demand GPU; among the best raw GPU value when capacity is available.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"H100 $3.99/h","key":"m0","label":"","primary":true},{"display":"A100 $2.79","key":"m1","label":"","primary":false},{"display":"40GB $1.99","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/together-ai-on-demand-dedicated-endpoints","slug":"together-ai-on-demand-dedicated-endpoints","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Together AI","provider":"Together AI","planName":"paid","kind":"Paid plan","description":"On-demand dedicated endpoints billed per minute: A100 40GB $2.40/hr, A100 80GB $2.59/hr, H100 80GB $3.36/hr (up to ~43% off legacy). Also serverless token APIs and clusters (H100 f","logo":"/logos/apis/together-ai.png","url":"https://www.together.ai/pricing","freeLimit":"Paid — H100 80GB $3.36/hr; A100 80GB $2.59/hr; A100 40GB $2.40/hr. On-demand dedicated endpoints billed per minute: A100 40GB $2.40/hr, A100 80GB $2.59/hr, H100 80GB $3.36/hr (up to ~43% off legacy). Also serverless token APIs and clusters (H100 from $5.49/hr).","bestFor":"A100 80GB at $2.59/hr and H100 at $3.36/hr for dedicated endpoints is competitive for guaranteed-capacity inference with no queue.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"H100 $3.36/h","key":"m0","label":"","primary":true},{"display":"A100 $2.59","key":"m1","label":"","primary":false},{"display":"40GB $2.40","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/inferless","slug":"inferless","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Inferless","provider":"Inferless","planName":"paid","kind":"Paid plan","description":"Serverless GPU inference billed per second with scale-to-zero; fractional/shared GPUs available. V2 shared A100 $0.000745/s; also dedicated GPUs (T4, A100, etc.).","logo":"/logos/inferless.png","url":"https://www.inferless.com/","freeLimit":"Paid — Shared A100 $0.000745/sec (~$2.68/hr). Serverless GPU inference billed per second with scale-to-zero; fractional/shared GPUs available. V2 shared A100 $0.000745/s; also dedicated GPUs (T4, A100, etc.).","bestFor":"Shared/fractional A100 at $0.000745/s plus fast cold starts; fractional GPU support means you only pay for the slice you need - good value for smaller","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"A100 ~$2.68","key":"m0","label":"","primary":true},{"display":"$745u/sec","key":"m1","label":"","primary":false},{"display":"per-sec","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/railway-hobby","slug":"railway-hobby","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Railway","provider":"Railway","planName":"paid","kind":"Paid plan","description":"Hobby: $5/mo with $5 usage credits; then ~$10/GB-RAM-month and ~$20/vCPU-month (billed per minute). Up to 48 vCPU / 48GB RAM per service.","logo":"/logos/railway-hobby.png","url":"https://railway.com/pricing","freeLimit":"Paid — $5/month (includes $5 usage credit). Hobby: $5/mo with $5 usage credits; then ~$10/GB-RAM-month and ~$20/vCPU-month (billed per minute). Up to 48 vCPU / 48GB RAM per service.","bestFor":"A 0.5 vCPU + 0.5GB service runs ~$0.80-$1.00/mo, so the $5 credit covers small always-on services; per-minute billing avoids paying for idle.","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":5,"valueText":"5.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$5/mo","key":"m0","label":"","primary":true},{"display":"$5 credit","key":"m1","label":"","primary":false},{"display":"$10/GB-mo","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟠 Usage-metered"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":5,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/fly-io-machines-performance-1x","slug":"fly-io-machines-performance-1x","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Fly.io Machines","provider":"Fly.io Machines","planName":"paid","kind":"Paid plan","description":"Dedicated-CPU micro-VM: performance-1x 2GB ~$0.0000120/s (~$31/mo always-on, ~$0.0431/hr); extra RAM ~$5/GB/month. Per-second billing, scale-to-zero when stopped.","logo":"/logos/fly-io-machines-performance-1x.png","url":"https://fly.io/docs/about/pricing/","freeLimit":"Paid — ~$31-32/month (performance-1x, 2GB). Dedicated-CPU micro-VM: performance-1x 2GB ~$0.0000120/s (~$31/mo always-on, ~$0.0431/hr); extra RAM ~$5/GB/month. Per-second billing, scale-to-zero when stopped.","bestFor":"A dedicated vCPU + 2GB always-on for ~$31/mo; per-second billing means intermittent workloads cost a fraction. Good value for consistent-performance a","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":31,"valueText":"31.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"~$31/mo","key":"m0","label":"","primary":true},{"display":"2GB","key":"m1","label":"","primary":false},{"display":"$.043/hr","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":31,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/vercel-functions-fluid-compute-pro","slug":"vercel-functions-fluid-compute-pro","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Vercel Functions","provider":"Vercel Functions","planName":"paid","kind":"Paid plan","description":"Pro seat $20/mo; Fluid compute billed as Active CPU $0.128/hr and Provisioned Memory $0.0106/GB-hour, plus invocations. Pro includes invocation/CPU allowances.","logo":"/logos/vercel-functions-fluid-compute-pro.png","url":"https://vercel.com/docs/functions/usage-and-pricing","freeLimit":"Paid — $20/month (Pro) + $0.128/Active-CPU-hour, $0.0106/GB-hour. Pro seat $20/mo; Fluid compute billed as Active CPU $0.128/hr and Provisioned Memory $0.0106/GB-hour, plus invocations. Pro includes invocation/CPU allowances.","bestFor":"~7.8 Active-CPU-hours per $1 ($0.128/hr); cheaper than legacy GB-s billing because you pay CPU only while actively computing (not during I/O wait).","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":20,"valueText":"20.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$20/mo","key":"m0","label":"","primary":true},{"display":"$.128 CPUh","key":"m1","label":"","primary":false},{"display":"$.0106/GBh","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":20,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/hugging-face-inference-endpoints","slug":"hugging-face-inference-endpoints","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Hugging Face Inference Endpoints","provider":"Hugging Face Inference Endpoints","planName":"paid","kind":"Paid plan","description":"Dedicated autoscaling inference billed per minute of uptime. CPU from $0.032/core/hr; GPUs T4 $0.50/hr, L4 $0.80/hr, A100 and up to ~$45/hr; can scale to zero.","logo":"/logos/hugging-face-inference-endpoints.png","url":"https://huggingface.co/docs/inference-endpoints/en/pricing","freeLimit":"Paid — From $0.032/CPU-core-hr; GPUs from $0.50/hr (T4). Dedicated autoscaling inference billed per minute of uptime. CPU from $0.032/core/hr; GPUs T4 $0.50/hr, L4 $0.80/hr, A100 and up to ~$45/hr; can scale to zero.","bestFor":"~2 T4 GPU-hours per $1 ($0.50/hr); per-minute billing and scale-to-zero make it efficient for spiky inference. CPU endpoints extremely cheap for light","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$.032/coreh","key":"m0","label":"","primary":true},{"display":"T4 $0.50/hr","key":"m1","label":"","primary":false},{"display":"per-min","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/koyeb-pro-serverless-compute","slug":"koyeb-pro-serverless-compute","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Koyeb","provider":"Koyeb","planName":"paid","kind":"Paid plan","description":"Pro $29/mo includes $10 compute; serverless CPU/GPU per second with scale-to-zero. GPU examples: L4 $0.70/hr, L40S $1.20/hr, A100 80GB $1.60/hr, H100 $2.50/hr, B200 $5.50/hr.","logo":"/logos/koyeb.png","url":"https://www.koyeb.com/pricing","freeLimit":"Paid — $29/month (Pro, $10 compute incl.) + per-second compute. Pro $29/mo includes $10 compute; serverless CPU/GPU per second with scale-to-zero. GPU examples: L4 $0.70/hr, L40S $1.20/hr, A100 80GB $1.60/hr, H100 $2.50/hr, B200 $5.50/hr.","bestFor":"A100 80GB at $1.60/hr is among the cheapest serverless A100 rates with scale-to-zero; per-second billing maximizes value for bursty workloads.","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":29,"valueText":"29.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$29/mo","key":"m0","label":"","primary":true},{"display":"$10 incl","key":"m1","label":"","primary":false},{"display":"per-sec","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Never sleeps"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.8,"unit":"score","value":4.8,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":9,"unit":"score","value":9},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.8,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":29,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/render-web-service-starter","slug":"render-web-service-starter","category":"serverless-functions","access":"conditional","accessBucket":"free","name":"Render","provider":"Render","planName":"starter","kind":"Paid plan","description":"Starter web service: 0.5 CPU / 512MB RAM at $7/mo (~$0.05/hr); Standard 1 CPU / 2GB at $32/mo (~$0.20/hr). Free tier of 750 instance-hours/month per workspace.","logo":"/logos/render-starter.png","url":"https://render.com/pricing","freeLimit":"Free tier of 750 instance-hours/month per workspace; then Starter web service $7/month (0.5 CPU / 512MB RAM, ~$0.05/hr), Standard 1 CPU / 2GB at $32/mo (~$0.20/hr).","bestFor":"Predictable flat per-instance pricing; a small always-on service for $7/mo. Workflows compute also at $0.05/hr (starter) per-hour.","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":7,"valueText":"7.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"freemium","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"freemium","keyMetrics":[{"display":"$7/mo","key":"m0","label":"","primary":true},{"display":"Std $32/mo","key":"m1","label":"","primary":false},{"display":"1CPU/2GB","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Flat rate","🟢 Never sleeps"],"valueModel":{"bucket":"freemium","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.9,"unit":"score","value":4.9,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"$/mo","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.8,"unit":"score","value":8.8},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8.8,"unit":"score","value":8.8}],"required_axis_count":3,"score":4.9,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"usd_month","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":7,"pricingModel":"freemium"},"subtop":"renewable"}},{"id":"serverless-functions/cloudflare-workers-for-platforms-paid","slug":"cloudflare-workers-for-platforms-paid","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Cloudflare Workers for Platforms","provider":"Cloudflare Workers for Platforms","planName":"paid","kind":"Paid plan","description":"Multi-tenant Workers platform for deploying customer/user code in isolated dispatch namespaces; $25/mo base plus underlying Workers usage. Run untrusted/agent code per-tenant at th","logo":"/logos/cloudflare-workers-for-platforms-paid.png","url":"https://developers.cloudflare.com/cloudflare-for-platforms/workers-for-platforms/reference/pricing/","freeLimit":"Paid — $25/month. Multi-tenant Workers platform for deploying customer/user code in isolated dispatch namespaces; $25/mo base plus underlying Workers usage. Run untrusted/agent code per-tenant at the edge.","bestFor":"At $25/mo plus standard Workers rates, you get unlimited per-tenant isolated functions - strong value for platforms running many users' or agents' cod","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":25,"valueText":"25.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$25/mo","key":"m0","label":"","primary":true},{"display":"multi-tenant","key":"m1","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Unlimited","🟠 Usage-metered"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":25,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/deno-deploy-pro","slug":"deno-deploy-pro","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Deno Deploy","provider":"Deno Deploy","planName":"paid","kind":"Paid plan","description":"Pro bundles ~5M included HTTP requests, 200GB egress, and CPU-hours, with overage egress $0.50/GB; per-org spend limits. Edge JS/TS (Deno) runtime with built-in KV.","logo":"/logos/deno-deploy-pro.png","url":"https://deno.com/deploy/pricing/","freeLimit":"Paid — $20/month. Pro bundles ~5M included HTTP requests, 200GB egress, and CPU-hours, with overage egress $0.50/GB; per-org spend limits. Edge JS/TS (Deno) runtime with built-in KV.","bestFor":"~250,000 requests per $1 of the $20 base (5M / $20). Strong for edge TypeScript apps; egress overage pricier than Cloudflare.","caveats":"","offerFacts":[{"key":"offered_price_usd","valueNumber":20,"valueText":"20.0","origin":"value_model_raw"},{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"$20/mo","key":"m0","label":"","primary":true},{"display":"~5M req","key":"m1","label":"","primary":false},{"display":"200GB out","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"offered_price_usd":20,"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/replicate","slug":"replicate","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Replicate","provider":"Replicate","planName":"paid","kind":"Paid plan","description":"Pay-per-second by hardware: CPU $0.000025/s, T4 $0.000225/s ($0.81/hr), L40S $3.51/hr, A100 80GB $5.04/hr, H100 $5.49/hr; multi-GPU available. Public models billed only for active ","logo":"/logos/apis/replicate.png","url":"https://replicate.com/pricing","freeLimit":"Paid — T4 $0.000225/s; A100 80GB $0.0014/s; H100 $0.001525/s. Pay-per-second by hardware: CPU $0.000025/s, T4 $0.000225/s ($0.81/hr), L40S $3.51/hr, A100 80GB $5.04/hr, H100 $5.49/hr; multi-GPU available. Public models billed only for active time.","bestFor":"T4 at $0.81/hr is cheap for light inference; A100/H100 rates higher than RunPod/Modal because private-model billing includes cold-start/idle. Best val","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"T4 $.81/hr","key":"m0","label":"","primary":true},{"display":"A100 $1.4m/s","key":"m1","label":"","primary":false},{"display":"H100 $1.5m","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/baseten","slug":"baseten","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Baseten","provider":"Baseten","planName":"paid","kind":"Paid plan","description":"Per-minute, idle-excluded billing on dedicated replicas; A100 80GB ~$4/hr, H100 MIG ~$0.0625/min, H100 80GB ~$6.50/hr (also $9.984/hr SXM tier). Autoscaling model deployments.","logo":"/logos/baseten.png","url":"https://www.baseten.co/pricing/","freeLimit":"Paid — A100 80GB ~$0.06667/min (~$4/hr); H100 ~$6.50/hr. Per-minute, idle-excluded billing on dedicated replicas; A100 80GB ~$4/hr, H100 MIG ~$0.0625/min, H100 80GB ~$6.50/hr (also $9.984/hr SXM tier). Autoscaling model deployments.","bestFor":"Per-minute billing with idle excluded improves value vs per-replica-hour platforms; rates above raw rental buy production reliability and autoscaling.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"A100 ~$4/hr","key":"m0","label":"","primary":true},{"display":"H100 $6.50/hr","key":"m1","label":"","primary":false},{"display":"per-min","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/together-ai-gpu-clusters","slug":"together-ai-gpu-clusters","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Together AI GPU Clusters","provider":"Together AI GPU Clusters","planName":"paid","kind":"Paid plan","description":"On-demand and reserved GPU clusters (8-256 GPUs): H100 SXM 80GB $5.49/hr on-demand, from $3.99/hr reserved; H200 and B200 also available. High-speed interconnect for distributed tr","logo":"/logos/together-ai-gpu-clusters.png","url":"https://www.together.ai/gpu-clusters","freeLimit":"Paid — H100 SXM from $3.99/GPU/hr (on-demand $5.49/hr). On-demand and reserved GPU clusters (8-256 GPUs): H100 SXM 80GB $5.49/hr on-demand, from $3.99/hr reserved; H200 and B200 also available. High-speed interconnect for distributed training/inference.","bestFor":"Reserved H100 from $3.99/hr with cluster-grade networking is solid for multi-GPU jobs; on-demand $5.49/hr buys instant scaling without commitment.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"H100 $3.99/hr","key":"m0","label":"","primary":true},{"display":"OD $5.49/hr","key":"m1","label":"","primary":false},{"display":"8-256 GPU","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}},{"id":"serverless-functions/replicate-multi-gpu-h100-cluster","slug":"replicate-multi-gpu-h100-cluster","category":"serverless-functions","access":"conditional","accessBucket":"paid","tier":"paid","name":"Replicate Multi-GPU","provider":"Replicate Multi-GPU","planName":"paid","kind":"Paid plan","description":"Per-second multi-GPU hardware: 2x A100 80GB $10.08/hr, 4x A100 $20.16/hr, 8x A100 $40.32/hr, 2x H100 $10.98/hr, 8x H100 $43.92/hr, L40S multi up to 8x $28.08/hr.","logo":"/logos/replicate-multi-gpu-h100-cluster.png","url":"https://replicate.com/pricing","freeLimit":"Paid — 8x H100 $0.012200/s ($43.92/hr); 2x A100 $0.0028/s. Per-second multi-GPU hardware: 2x A100 80GB $10.08/hr, 4x A100 $20.16/hr, 8x A100 $40.32/hr, 2x H100 $10.98/hr, 8x H100 $43.92/hr, L40S multi up to 8x $28.08/hr.","bestFor":"Linear per-GPU scaling (e.g. $5.49/hr per H100) with per-second billing; good for running big models on-demand without negotiating clusters.","caveats":"","offerFacts":[{"key":"pricingModel","valueText":"paid","origin":"value_model_raw"}],"tags":["paid","value","serverless-functions"],"pricingModel":"paid","keyMetrics":[{"display":"8xH100 $44h","key":"m0","label":"","primary":true},{"display":"8xA100 $40h","key":"m1","label":"","primary":false},{"display":"$12.2m/s","key":"m2","label":"","primary":false}],"verified":true,"timestamp":"2026-06-18T00:00:00Z","verifiedAt":"2026-06-18","badges":["🔴 Paid only","🟢 Commercial OK","🟢 Embeddings/RAG"],"valueModel":{"bucket":"paid","calc":{"rating_axis_count":3,"ratings":[{"a":1,"b":0,"formula":"weighted_components_v1","key":"main","label":"Overall","provenance":"weighted_capacity_fit_access","score":4.6,"unit":"score","value":4.6,"weights":{"access":0.2,"capacity":0.5,"fit":0.3}},{"a":2,"anchor":0.01,"b":5,"formula":"fixed_log_v1","key":"capacity","label":"Invocation value","provenance":"policy_default_missing_provider_capacity","score":1,"unit":"×","value":0.01},{"a":1,"b":0,"formula":"policy_score_v1","key":"fit","label":"Runtime fit","provenance":"derived_from_limitations_badges_and_access_terms","score":8.2,"unit":"score","value":8.2},{"a":1,"b":0,"formula":"policy_score_v1","key":"access","label":"Deploy friction","provenance":"derived_from_limitations_badges_and_access_terms","score":8,"unit":"score","value":8}],"required_axis_count":3,"score":4.6,"score_formula":"weighted_components_v1","value":0.01,"value_kind":"ratio","value_provenance":"policy_default_missing_provider_capacity"},"commercial":{"source":"inferred","value":true},"generated_by":"scripts/score-infra-value.py","quantities":[],"raw":{"pricingModel":"paid"},"subtop":"renewable"}}],"categories":[{"id":"llm-providers","label":"LLM","description":"Free or BYOK model APIs"},{"id":"embeddings","label":"Embeddings","description":"Embedding, reranker, and vector search services"},{"id":"databases","label":"DB","description":"Structured app state"},{"id":"app-platforms","label":"Apps","description":"Deploy full apps & services"},{"id":"agent-deployment","label":"Agents","description":"Deploy & run AI agents"},{"id":"serverless-functions","label":"Functions","description":"Run functions, edge & jobs"},{"id":"schedulers","label":"Schedulers","description":"Cron jobs & recurring tasks"},{"id":"storage","label":"Storage","description":"Files, datasets, bundles"},{"id":"workspaces","label":"Workspace","description":"Notion-like user surfaces"},{"id":"domains","label":"Domains","description":"Free hostnames and DNS"}],"selectedCategory":"serverless-functions","showSidebar":false,"categoryCounts":{"agent-deployment":88,"app-platforms":55,"databases":61,"domains":64,"embeddings":11,"llm-providers":106,"schedulers":50,"serverless-functions":58,"storage":56,"workspaces":70},"softwareCategoryCounts":{"agent-deployment":40,"agent-frameworks":39,"agents":41,"browser-computer-use":28,"chat-ui":27,"data-extraction":31,"fine-tuning":36,"guardrails-security":27,"inference-serving":33,"llm-gateways":37,"mcp-tools":30,"observability-evals":36,"orchestration":20,"rag-vector-memory":39,"sandboxes":32,"voice-realtime":41,"web-crawl-search":40},"apiCategoryCounts":{"ai-ml":34,"business":31,"communication":37,"data-open":24,"dev-tools":31,"entertainment":37,"finance-crypto":35,"food-travel":26,"geo-maps":36,"gov-science":27,"media-images":31,"news-media":32,"search-web":35,"social-comm":26,"sports":34,"weather":30},"agentToolCategoryCounts":{"mcp-cloud":26,"mcp-data":32,"mcp-dev":33,"mcp-other":10,"mcp-productivity":33,"mcp-search":34,"plugins":19,"prompt-libraries":24,"registries":16,"resources":25,"skills":26,"toolkits":22},"totalOfferCount":2002,"documentationSections":[{"id":"methodology","name":"Methodology","items":[{"id":"generosity-measurement","title":"How ratings are computed","href":"/documentation/generosity-measurement"},{"id":"free-tier-conditions","title":"Free-Tier Conditions & Limitations","href":"/documentation/free-tier-conditions"},{"id":"badges-and-labels","title":"Badges, labels & markers","href":"/documentation/badges-and-labels"}]}],"recipeSections":[{"id":"personal-assistant","name":"Personal Assistant","items":[{"id":"archive/notion-upstash-telegram-morning-agent","title":"Morning Notion Task Digest to Telegram","href":"/recipes/archive/notion-upstash-telegram-morning-agent"},{"id":"archive/theme0-langgraph-gmail-email-assistant","title":"Human-in-the-Loop Gmail Email Assistant","href":"/recipes/archive/theme0-langgraph-gmail-email-assistant"},{"id":"archive/theme0-google-calendar-mcp-agent","title":"Natural-Language Google Calendar Agent (MCP)","href":"/recipes/archive/theme0-google-calendar-mcp-agent"},{"id":"archive/theme0-second-brain-notes-rag","title":"Second-Brain Notes Assistant (RAG over your vault)","href":"/recipes/archive/theme0-second-brain-notes-rag"},{"id":"archive/theme0-crewai-daily-briefing-crew","title":"Multi-Agent Daily Briefing Crew","href":"/recipes/archive/theme0-crewai-daily-briefing-crew"}]},{"id":"scheduled-automation","name":"Scheduled Automation","items":[{"id":"archive/theme6-scheduled-log-anomaly-sentinel","title":"Scheduled Log-Anomaly Sentinel with Vector Search","href":"/recipes/archive/theme6-scheduled-log-anomaly-sentinel"},{"id":"archive/theme5-triggerdev-youtube-content-digest","title":"Scheduled YouTube-to-Email Content Digest","href":"/recipes/archive/theme5-triggerdev-youtube-content-digest"},{"id":"archive/theme5-firecrawl-competitor-newsletter","title":"Curated Competitor-News Newsletter from Live Crawls","href":"/recipes/archive/theme5-firecrawl-competitor-newsletter"},{"id":"archive/theme3-trigger-dev-hacker-news-scrape-summary","title":"Scheduled Hacker News Scrape, Summarize, and Email Digest","href":"/recipes/archive/theme3-trigger-dev-hacker-news-scrape-summary"},{"id":"archive/theme3-firecrawl-crawl-docs-upstash-vector-rag","title":"Crawl a Documentation Site into an Upstash Vector RAG Index","href":"/recipes/archive/theme3-firecrawl-crawl-docs-upstash-vector-rag"},{"id":"archive/theme3-langchain-incremental-indexing-pgvector-sync","title":"Incrementally Sync a Source into a pgvector Index Without Re-Embedding","href":"/recipes/archive/theme3-langchain-incremental-indexing-pgvector-sync"},{"id":"archive/theme0-firecrawl-news-digest-email","title":"Daily Web Digest to Your Inbox (Firecrawl)","href":"/recipes/archive/theme0-firecrawl-news-digest-email"},{"id":"archive/theme0-triggerdev-per-user-morning-briefing","title":"Per-User Morning Briefing on Trigger.dev","href":"/recipes/archive/theme0-triggerdev-per-user-morning-briefing"},{"id":"archive/theme1-triggerdev-scheduled-scrape-summary-digest","title":"Scheduled Scrape-and-Summarize Digest with Trigger.dev","href":"/recipes/archive/theme1-triggerdev-scheduled-scrape-summary-digest"}]},{"id":"event-automation","name":"Event Automation","items":[{"id":"archive/theme6-autonomous-incident-triage-agent","title":"Autonomous Incident Triage Agent for On-Call SREs","href":"/recipes/archive/theme6-autonomous-incident-triage-agent"},{"id":"archive/theme6-terraform-pr-review-agent-mcp","title":"Terraform PR Agent with the HashiCorp MCP Server","href":"/recipes/archive/theme6-terraform-pr-review-agent-mcp"},{"id":"archive/theme2-langgraph-pr-review-agent","title":"Automated PR Review Agent with LangGraph","href":"/recipes/archive/theme2-langgraph-pr-review-agent"},{"id":"archive/theme5-langgraph-social-media-post-agent","title":"URL to Twitter & LinkedIn Posts with Human Review","href":"/recipes/archive/theme5-langgraph-social-media-post-agent"},{"id":"archive/theme4-crewai-multi-agent-ticket-triage","title":"Multi-Agent Support Ticket Triage with CrewAI","href":"/recipes/archive/theme4-crewai-multi-agent-ticket-triage"},{"id":"archive/theme4-zendesk-webhook-triage-agent","title":"Zendesk Ticket Triage Agent: Classify, Search the KB, Route to Slack","href":"/recipes/archive/theme4-zendesk-webhook-triage-agent"},{"id":"archive/theme3-langchain-invoice-email-to-postgres","title":"Extract Invoice and Receipt Data from Email into Postgres","href":"/recipes/archive/theme3-langchain-invoice-email-to-postgres"},{"id":"archive/theme2-trigger-dev-github-issue-triage","title":"GitHub Issue Triage and PR Review on Trigger.dev","href":"/recipes/archive/theme2-trigger-dev-github-issue-triage"},{"id":"archive/theme1-firecrawl-monitor-competitor-watch-agent","title":"Competitor Web-Change Watch Agent with Firecrawl Monitor","href":"/recipes/archive/theme1-firecrawl-monitor-competitor-watch-agent"}]},{"id":"worker","name":"Worker","items":[{"id":"cloudflare-free-app-stack","title":"Cloudflare Free App Stack","href":"/recipes/cloudflare-free-app-stack"},{"id":"cloudflare-ai-gateway-free-llm-fallback","title":"Cloudflare AI Gateway Free LLM Fallback Chain","href":"/recipes/cloudflare-ai-gateway-free-llm-fallback"},{"id":"archive/theme6-log-clustering-triage-agent","title":"Log Clustering and Triage Agent over MCP","href":"/recipes/archive/theme6-log-clustering-triage-agent"},{"id":"vercel-agent-ai-gateway-central-hub","title":"Vercel Agent AI Gateway Central Hub","href":"/recipes/vercel-agent-ai-gateway-central-hub"},{"id":"archive/theme7-orchestrator-workers-fact-checker","title":"Orchestrator-Workers News Fact-Checker","href":"/recipes/archive/theme7-orchestrator-workers-fact-checker"},{"id":"archive/theme4-firecrawl-docs-to-knowledge-base-agent","title":"Turn a Documentation Site into a Support Agent with Firecrawl and LangGraph","href":"/recipes/archive/theme4-firecrawl-docs-to-knowledge-base-agent"},{"id":"archive/theme5-triggerdev-youtube-content-digest","title":"Scheduled YouTube-to-Email Content Digest","href":"/recipes/archive/theme5-triggerdev-youtube-content-digest"},{"id":"archive/theme7-crewai-hierarchical-content-crew","title":"CrewAI Hierarchical Content Crew","href":"/recipes/archive/theme7-crewai-hierarchical-content-crew"},{"id":"archive/theme5-firecrawl-competitor-newsletter","title":"Curated Competitor-News Newsletter from Live Crawls","href":"/recipes/archive/theme5-firecrawl-competitor-newsletter"},{"id":"archive/theme7-parallel-subagent-research-swarm","title":"Parallel Subagent Research Swarm","href":"/recipes/archive/theme7-parallel-subagent-research-swarm"},{"id":"archive/theme5-content-repurposing-worker","title":"Repurpose One Blog Post into a Week of Channels","href":"/recipes/archive/theme5-content-repurposing-worker"},{"id":"archive/theme2-cover-agent-test-generation","title":"Coverage-Driven Unit Test Generation with Cover-Agent","href":"/recipes/archive/theme2-cover-agent-test-generation"},{"id":"archive/theme3-trigger-dev-hacker-news-scrape-summary","title":"Scheduled Hacker News Scrape, Summarize, and Email Digest","href":"/recipes/archive/theme3-trigger-dev-hacker-news-scrape-summary"},{"id":"archive/theme3-crewai-multi-agent-lead-scraper","title":"Multi-Agent Lead Scraper that Outputs a Structured Leads Table","href":"/recipes/archive/theme3-crewai-multi-agent-lead-scraper"},{"id":"archive/theme3-langchain-invoice-email-to-postgres","title":"Extract Invoice and Receipt Data from Email into Postgres","href":"/recipes/archive/theme3-langchain-invoice-email-to-postgres"},{"id":"archive/theme0-firecrawl-news-digest-email","title":"Daily Web Digest to Your Inbox (Firecrawl)","href":"/recipes/archive/theme0-firecrawl-news-digest-email"},{"id":"archive/theme1-crewai-competitive-analysis-crew","title":"Competitive Analysis Crew that Outputs a Strategic Report","href":"/recipes/archive/theme1-crewai-competitive-analysis-crew"},{"id":"archive/theme1-triggerdev-scheduled-scrape-summary-digest","title":"Scheduled Scrape-and-Summarize Digest with Trigger.dev","href":"/recipes/archive/theme1-triggerdev-scheduled-scrape-summary-digest"}]},{"id":"workflow","name":"Workflow","items":[{"id":"cloudflare-free-app-stack","title":"Cloudflare Free App Stack","href":"/recipes/cloudflare-free-app-stack"},{"id":"cloudflare-ai-gateway-free-llm-fallback","title":"Cloudflare AI Gateway Free LLM Fallback Chain","href":"/recipes/cloudflare-ai-gateway-free-llm-fallback"},{"id":"vercel-agent-ai-gateway-central-hub","title":"Vercel Agent AI Gateway Central Hub","href":"/recipes/vercel-agent-ai-gateway-central-hub"},{"id":"archive/theme6-natural-language-iac-generator","title":"Natural-Language Infrastructure-as-Code Generator","href":"/recipes/archive/theme6-natural-language-iac-generator"},{"id":"archive/theme6-terraform-pr-review-agent-mcp","title":"Terraform PR Agent with the HashiCorp MCP Server","href":"/recipes/archive/theme6-terraform-pr-review-agent-mcp"},{"id":"archive/theme2-langgraph-pr-review-agent","title":"Automated PR Review Agent with LangGraph","href":"/recipes/archive/theme2-langgraph-pr-review-agent"},{"id":"archive/theme7-orchestrator-workers-fact-checker","title":"Orchestrator-Workers News Fact-Checker","href":"/recipes/archive/theme7-orchestrator-workers-fact-checker"},{"id":"archive/theme7-evaluator-optimizer-translation-loop","title":"Evaluator-Optimizer Translation Loop","href":"/recipes/archive/theme7-evaluator-optimizer-translation-loop"},{"id":"archive/theme5-crewai-seo-blog-editorial-crew","title":"Multi-Agent SEO Blog Pipeline that Publishes to Ghost","href":"/recipes/archive/theme5-crewai-seo-blog-editorial-crew"},{"id":"archive/theme7-langgraph-supervisor-research-desk","title":"LangGraph Supervisor Research Desk","href":"/recipes/archive/theme7-langgraph-supervisor-research-desk"},{"id":"archive/theme4-crewai-multi-agent-ticket-triage","title":"Multi-Agent Support Ticket Triage with CrewAI","href":"/recipes/archive/theme4-crewai-multi-agent-ticket-triage"},{"id":"archive/theme5-qdrant-brand-aware-content-rag","title":"Brand-Aware Content Generator with RAG","href":"/recipes/archive/theme5-qdrant-brand-aware-content-rag"},{"id":"archive/theme4-inngest-durable-rag-answer-workflow","title":"Durable RAG Answer Workflow with Inngest","href":"/recipes/archive/theme4-inngest-durable-rag-answer-workflow"},{"id":"archive/theme7-llamaindex-agentworkflow-report-swarm","title":"LlamaIndex AgentWorkflow Report Swarm","href":"/recipes/archive/theme7-llamaindex-agentworkflow-report-swarm"},{"id":"archive/theme5-content-repurposing-worker","title":"Repurpose One Blog Post into a Week of Channels","href":"/recipes/archive/theme5-content-repurposing-worker"},{"id":"archive/theme2-cover-agent-test-generation","title":"Coverage-Driven Unit Test Generation with Cover-Agent","href":"/recipes/archive/theme2-cover-agent-test-generation"},{"id":"archive/theme3-llamaparse-pdf-manuals-qdrant-hybrid-search","title":"Parse Complex PDF Manuals into a Qdrant Hybrid-Search Knowledge Base","href":"/recipes/archive/theme3-llamaparse-pdf-manuals-qdrant-hybrid-search"},{"id":"archive/theme2-openhands-dependency-migration-agent","title":"Autonomous Dependency Migration Agent with OpenHands","href":"/recipes/archive/theme2-openhands-dependency-migration-agent"},{"id":"archive/theme0-crewai-daily-briefing-crew","title":"Multi-Agent Daily Briefing Crew","href":"/recipes/archive/theme0-crewai-daily-briefing-crew"},{"id":"archive/theme2-trigger-dev-github-issue-triage","title":"GitHub Issue Triage and PR Review on Trigger.dev","href":"/recipes/archive/theme2-trigger-dev-github-issue-triage"},{"id":"archive/theme1-firecrawl-deep-research-agent","title":"Deep Research Agent with Firecrawl and an Agent SDK","href":"/recipes/archive/theme1-firecrawl-deep-research-agent"},{"id":"archive/theme2-firecrawl-qdrant-migration-assistant","title":"Live-Docs Migration Assistant with Firecrawl and Qdrant","href":"/recipes/archive/theme2-firecrawl-qdrant-migration-assistant"},{"id":"archive/theme0-triggerdev-per-user-morning-briefing","title":"Per-User Morning Briefing on Trigger.dev","href":"/recipes/archive/theme0-triggerdev-per-user-morning-briefing"},{"id":"archive/theme1-langgraph-exa-deep-research-pipeline","title":"Auditable Deep Research Pipeline with LangGraph and Exa","href":"/recipes/archive/theme1-langgraph-exa-deep-research-pipeline"}]},{"id":"agent-runtime","name":"Agent Runtime","items":[{"id":"cloudflare-free-app-stack","title":"Cloudflare Free App Stack","href":"/recipes/cloudflare-free-app-stack"},{"id":"archive/theme6-autonomous-incident-triage-agent","title":"Autonomous Incident Triage Agent for On-Call SREs","href":"/recipes/archive/theme6-autonomous-incident-triage-agent"},{"id":"cloudflare-ai-gateway-free-llm-fallback","title":"Cloudflare AI Gateway Free LLM Fallback Chain","href":"/recipes/cloudflare-ai-gateway-free-llm-fallback"},{"id":"vercel-agent-ai-gateway-central-hub","title":"Vercel Agent AI Gateway Central Hub","href":"/recipes/vercel-agent-ai-gateway-central-hub"},{"id":"archive/theme6-natural-language-iac-generator","title":"Natural-Language Infrastructure-as-Code Generator","href":"/recipes/archive/theme6-natural-language-iac-generator"},{"id":"archive/theme6-kubernetes-sre-helper-agent","title":"Kubernetes SRE Helper Agent for Multi-Cluster Toil","href":"/recipes/archive/theme6-kubernetes-sre-helper-agent"},{"id":"archive/theme0-langgraph-gmail-email-assistant","title":"Human-in-the-Loop Gmail Email Assistant","href":"/recipes/archive/theme0-langgraph-gmail-email-assistant"},{"id":"archive/theme5-langgraph-social-media-post-agent","title":"URL to Twitter & LinkedIn Posts with Human Review","href":"/recipes/archive/theme5-langgraph-social-media-post-agent"},{"id":"archive/theme7-langgraph-supervisor-research-desk","title":"LangGraph Supervisor Research Desk","href":"/recipes/archive/theme7-langgraph-supervisor-research-desk"},{"id":"archive/theme7-crewai-hierarchical-content-crew","title":"CrewAI Hierarchical Content Crew","href":"/recipes/archive/theme7-crewai-hierarchical-content-crew"},{"id":"archive/theme4-zendesk-webhook-triage-agent","title":"Zendesk Ticket Triage Agent: Classify, Search the KB, Route to Slack","href":"/recipes/archive/theme4-zendesk-webhook-triage-agent"},{"id":"archive/theme7-llamaindex-agentworkflow-report-swarm","title":"LlamaIndex AgentWorkflow Report Swarm","href":"/recipes/archive/theme7-llamaindex-agentworkflow-report-swarm"},{"id":"archive/theme4-llamaindex-document-agents-support","title":"Document Agents over Manuals and FAQs with LlamaIndex","href":"/recipes/archive/theme4-llamaindex-document-agents-support"},{"id":"archive/theme7-parallel-subagent-research-swarm","title":"Parallel Subagent Research Swarm","href":"/recipes/archive/theme7-parallel-subagent-research-swarm"},{"id":"archive/theme2-llamaindex-codebase-qa","title":"Chat With Your Codebase Using LlamaIndex RAG","href":"/recipes/archive/theme2-llamaindex-codebase-qa"},{"id":"archive/theme0-google-calendar-mcp-agent","title":"Natural-Language Google Calendar Agent (MCP)","href":"/recipes/archive/theme0-google-calendar-mcp-agent"},{"id":"archive/theme3-crewai-multi-agent-lead-scraper","title":"Multi-Agent Lead Scraper that Outputs a Structured Leads Table","href":"/recipes/archive/theme3-crewai-multi-agent-lead-scraper"},{"id":"archive/theme2-openhands-dependency-migration-agent","title":"Autonomous Dependency Migration Agent with OpenHands","href":"/recipes/archive/theme2-openhands-dependency-migration-agent"},{"id":"archive/theme1-firecrawl-deep-research-agent","title":"Deep Research Agent with Firecrawl and an Agent SDK","href":"/recipes/archive/theme1-firecrawl-deep-research-agent"},{"id":"archive/theme1-langgraph-exa-deep-research-pipeline","title":"Auditable Deep Research Pipeline with LangGraph and Exa","href":"/recipes/archive/theme1-langgraph-exa-deep-research-pipeline"},{"id":"archive/theme1-crewai-competitive-analysis-crew","title":"Competitive Analysis Crew that Outputs a Strategic Report","href":"/recipes/archive/theme1-crewai-competitive-analysis-crew"},{"id":"archive/theme1-searxng-private-web-research-agent","title":"Private Web Research Agent with Self-Hosted SearXNG","href":"/recipes/archive/theme1-searxng-private-web-research-agent"}]},{"id":"notification","name":"Notification","items":[{"id":"archive/theme6-kubernetes-sre-helper-agent","title":"Kubernetes SRE Helper Agent for Multi-Cluster Toil","href":"/recipes/archive/theme6-kubernetes-sre-helper-agent"},{"id":"archive/theme0-firecrawl-news-digest-email","title":"Daily Web Digest to Your Inbox (Firecrawl)","href":"/recipes/archive/theme0-firecrawl-news-digest-email"},{"id":"archive/theme0-triggerdev-per-user-morning-briefing","title":"Per-User Morning Briefing on Trigger.dev","href":"/recipes/archive/theme0-triggerdev-per-user-morning-briefing"},{"id":"archive/theme1-firecrawl-monitor-competitor-watch-agent","title":"Competitor Web-Change Watch Agent with Firecrawl Monitor","href":"/recipes/archive/theme1-firecrawl-monitor-competitor-watch-agent"}]},{"id":"knowledge-base","name":"Knowledge Base","items":[{"id":"archive/theme6-log-clustering-triage-agent","title":"Log Clustering and Triage Agent over MCP","href":"/recipes/archive/theme6-log-clustering-triage-agent"},{"id":"archive/theme6-scheduled-log-anomaly-sentinel","title":"Scheduled Log-Anomaly Sentinel with Vector Search","href":"/recipes/archive/theme6-scheduled-log-anomaly-sentinel"},{"id":"archive/theme4-langgraph-qdrant-hybrid-support-chatbot","title":"Hybrid-Search Customer Support Chatbot with LangGraph and Qdrant","href":"/recipes/archive/theme4-langgraph-qdrant-hybrid-support-chatbot"},{"id":"archive/theme4-firecrawl-docs-to-knowledge-base-agent","title":"Turn a Documentation Site into a Support Agent with Firecrawl and LangGraph","href":"/recipes/archive/theme4-firecrawl-docs-to-knowledge-base-agent"},{"id":"archive/theme5-qdrant-brand-aware-content-rag","title":"Brand-Aware Content Generator with RAG","href":"/recipes/archive/theme5-qdrant-brand-aware-content-rag"},{"id":"archive/theme4-llamaindex-document-agents-support","title":"Document Agents over Manuals and FAQs with LlamaIndex","href":"/recipes/archive/theme4-llamaindex-document-agents-support"},{"id":"archive/theme2-llamaindex-codebase-qa","title":"Chat With Your Codebase Using LlamaIndex RAG","href":"/recipes/archive/theme2-llamaindex-codebase-qa"},{"id":"archive/theme0-second-brain-notes-rag","title":"Second-Brain Notes Assistant (RAG over your vault)","href":"/recipes/archive/theme0-second-brain-notes-rag"},{"id":"archive/theme3-llamaparse-pdf-manuals-qdrant-hybrid-search","title":"Parse Complex PDF Manuals into a Qdrant Hybrid-Search Knowledge Base","href":"/recipes/archive/theme3-llamaparse-pdf-manuals-qdrant-hybrid-search"},{"id":"archive/theme3-firecrawl-crawl-docs-upstash-vector-rag","title":"Crawl a Documentation Site into an Upstash Vector RAG Index","href":"/recipes/archive/theme3-firecrawl-crawl-docs-upstash-vector-rag"},{"id":"archive/theme3-langchain-incremental-indexing-pgvector-sync","title":"Incrementally Sync a Source into a pgvector Index Without Re-Embedding","href":"/recipes/archive/theme3-langchain-incremental-indexing-pgvector-sync"},{"id":"archive/theme2-firecrawl-qdrant-migration-assistant","title":"Live-Docs Migration Assistant with Firecrawl and Qdrant","href":"/recipes/archive/theme2-firecrawl-qdrant-migration-assistant"},{"id":"archive/theme1-searxng-private-web-research-agent","title":"Private Web Research Agent with Self-Hosted SearXNG","href":"/recipes/archive/theme1-searxng-private-web-research-agent"}]}],"curatedHighlightsByCategory":{"llm-providers":{"collection":"infrastructure","category":"llm-providers","updatedAt":"2026-07-13","picks":[{"slug":"google-gemini","headline":"The default free frontier API","description":"The strongest standing free tier in the catalog: current Gemini frontier models, a real free quota that renews daily, and no credit card at any point. Tool calling, structured output, and a long context window are all on the free tier — the whole modern feature set, not a crippled demo. Note the two honest trade-offs: the API is Google-native REST rather than OpenAI-compatible, and free-tier prompts are used for model training with no opt-out. Rate limits vary by model and region, but for agents, prototypes, and side projects the free tier is genuinely workable as a daily driver. If you only configure one free LLM provider, make it this one.","whyChips":["No card","Frontier models","Model-specific limits"],"verified":{"verifiedAt":"2026-07-13","confidence":"high","note":"Official Google pricing confirms a free tier and training use; official rate-limit docs say RPM, TPM, and RPD vary by model/project and must be read from AI Studio. No blanket 5,000-request quota is claimed."},"offer":{"slug":"google-gemini","title":"Google AI Studio / Gemini API","provider":"Google AI Studio / Gemini API","planName":"freemium","logo":"/logos/gemini.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier exists; exact RPM, TPM, and RPD vary by model, project, region, and the live limits shown in Google AI Studio.","keyMetrics":[{"key":"model","label":"Best model","value":"50.2 AA","display":"50.2 AA"}],"url":"https://ai.google.dev/gemini-api/docs"}},{"slug":"groq","headline":"Absurd speed, free to start","description":"LPU inference makes open models (Llama, Qwen, GPT-OSS) respond at hundreds of tokens per second — interactive agents feel instant. The free plan has honest per-model RPM/RPD/TPM limits visible in the console, no card required. When latency is the feature, start here.","whyChips":["Fastest inference","No card"],"verified":{"verifiedAt":"2026-07-07","confidence":"high","note":"All five S2.5 points live-verified (console rate-limit docs, pricing, ToS, model list); no card; commercial OK."},"offer":{"slug":"groq","title":"GroqCloud","provider":"GroqCloud","planName":"freemium","logo":"/logos/groq.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan limits are shown per organization/model in Groq Console; measured as RPM/RPD/TPM/TPD.","keyMetrics":[{"key":"model","label":"Best model","value":"23.8 AA","display":"23.8 AA"},{"key":"tokens_per_minute","label":"Tokens/min","value":"6K T/min","display":"6K T/min"},{"key":"requests_per_day","label":"Requests/day","value":"14.4K R/day","display":"14.4K R/day"}],"url":"https://console.groq.com/"}},{"slug":"openrouter","headline":"27 free models behind one key","description":"One OpenAI-compatible key routes to every provider's zero-price models — 27 free routes at last count, including strong recent open-weights. Free-model limits are request-based (about 20 RPM / 50 per day without a top-up), which is plenty for evaluation and light agent duty. The single-integration hedge against any one provider changing its terms.","whyChips":["One key, many models","OpenAI-compatible"],"verified":{"verifiedAt":"2026-07-07","confidence":"high","note":"Live-verified: 27+ :free routes, 20 RPM / 50 RPD (1,000 RPD with $10 credit), no card, commercial OK; credit balance must stay at $0 or above."},"offer":{"slug":"openrouter","title":"OpenRouter","provider":"OpenRouter","planName":"freemium","logo":"/logos/apis/openrouter.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"27 zero-price OpenRouter routes; no-top-up free-model limits are request-based (20 RPM, 50 requests/day). Buying credits raises the daily limit.","keyMetrics":[{"key":"model","label":"Best model","value":"37.8 AA","display":"37.8 AA"},{"key":"requests_per_day","label":"Requests/day","value":"50 R/day","display":"50 R/day"},{"key":"requests_per_minute","label":"Requests/min","value":"20 R/min","display":"20 R/min"}],"url":"https://openrouter.ai/"}}],"hero":{"slug":"google-gemini","headline":"The default free frontier API","description":"The strongest standing free tier in the catalog: current Gemini frontier models, a real free quota that renews daily, and no credit card at any point. Tool calling, structured output, and a long context window are all on the free tier — the whole modern feature set, not a crippled demo. Note the two honest trade-offs: the API is Google-native REST rather than OpenAI-compatible, and free-tier prompts are used for model training with no opt-out. Rate limits vary by model and region, but for agents, prototypes, and side projects the free tier is genuinely workable as a daily driver. If you only configure one free LLM provider, make it this one.","whyChips":["No card","Frontier models","Model-specific limits"],"verified":{"verifiedAt":"2026-07-13","confidence":"high","note":"Official Google pricing confirms a free tier and training use; official rate-limit docs say RPM, TPM, and RPD vary by model/project and must be read from AI Studio. No blanket 5,000-request quota is claimed."},"offer":{"slug":"google-gemini","title":"Google AI Studio / Gemini API","provider":"Google AI Studio / Gemini API","planName":"freemium","logo":"/logos/gemini.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier exists; exact RPM, TPM, and RPD vary by model, project, region, and the live limits shown in Google AI Studio.","keyMetrics":[{"key":"model","label":"Best model","value":"50.2 AA","display":"50.2 AA"}],"url":"https://ai.google.dev/gemini-api/docs"}},"nominations":[{"slug":"mistral-ai","label":"Frontier free","description":"Frontier Mistral models (Medium 3.5, Large 3, Codestral) on a no-card Experiment tier that is ongoing, not a trial. Data training is on by default and the tier is framed for evaluation — flip the opt-out and it is the best EU-hosted free entry point.","verified":{"verifiedAt":"2026-07-10","confidence":"high","note":"Live-verified: ongoing no-card free tier, 10 models incl. frontier; exact limits login-gated; data-training default-on; commercial use requires paid Scale plan."},"icon":"medal","offer":{"slug":"mistral-ai","title":"Mistral AI","provider":"Mistral AI","planName":"freemium","logo":"/logos/mistral-ai.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free 'Experiment' plan, no credit card; public Mistral docs state limited free-mode rate limits and a token/month cap, but do not publish a numeric monthly token allowance. Prompts may be used to improve models.","keyMetrics":[{"key":"model","label":"Best model","value":"12.5 AA","display":"12.5 AA"}],"url":"https://console.mistral.ai/api-keys"}},{"slug":"ollama-cloud","label":"Privacy first","description":"A genuine $0 tier metered by GPU time rather than tokens, with frontier open models and a no-logging, no-training requirement on every partner. The only shortlist entry built around data privacy.","verified":{"verifiedAt":"2026-07-09","confidence":"high","note":"Live pricing page: $0 plan, GPU-time metering, 5h/7d limit resets, privacy/no-logging language; commercial allowed."},"icon":"medal","offer":{"slug":"ollama-cloud","title":"Ollama Cloud","provider":"Ollama Cloud","planName":"freemium","logo":"/logos/ollama-cloud.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free $0 plan includes access to cloud models with reported 'light usage' limits: session limits reset every 5 hours and weekly limits every 7 days; usage measured by GPU time, not tokens. Pro ($20/mo) and Max ($100/mo) raise limits 50x/250x.","keyMetrics":[{"key":"model","label":"Best model","value":"33.8 AA","display":"33.8 AA"}],"url":"https://ollama.com/cloud"}},{"slug":"fireworks-ai","label":"Model breadth","description":"34 models — DeepSeek V4, Llama 4, Qwen 3.6, Kimi K2, GLM-5.2 — behind one OpenAI-compatible key, with a $1 no-card signup credit that covers real evaluation. Production continues seamlessly as pay-as-you-go.","verified":{"verifiedAt":"2026-07-09","confidence":"high","note":"Live pricing page: $1 one-time signup credit, no card, instant key, OpenAI-compatible; commercial allowed; finite credit, not ongoing-free."},"icon":"medal","offer":{"slug":"fireworks-ai","title":"Fireworks AI","provider":"Fireworks AI","planName":"finite_credit","logo":"/logos/apis/fireworks-ai.png","pricingModel":"finite_credit","accessBucket":"conditional","freeLimitText":"$1 in free credits on signup for serverless text and vision models; pay-as-you-go thereafter. Finite, one-time.","keyMetrics":[{"key":"model","label":"Best model","value":"51.1 AA","display":"51.1 AA"},{"key":"requests_per_minute","label":"Requests/min","value":"10 R/min","display":"10 R/min"},{"key":"credit_amount_usd","label":"Credit","value":"1 USD","display":"1 USD"}],"url":"https://fireworks.ai/"}}],"paidPicks":[{"slug":"kimi-code-moderato","description":"A focused coding subscription for Kimi's long-context code models with OpenAI and Anthropic-compatible endpoints.","priceFact":"$19/mo","offer":{"slug":"kimi-code-moderato","title":"Kimi Code Moderato","provider":"Kimi Code Moderato","planName":"paid","logo":"/logos/kimi-code-moderato.png","pricingModel":"paid","accessBucket":"paid","freeLimitText":"Paid — $19/mo ($15/mo billed annually). Kimi K2.7 Code + K2.7 HighSpeed (~180 tok/s) + K2.6; ~300-1,200 requests/5h with weekly quota, up to 30 concurrent; OpenAI- AND Anthropic-compatible endpoints (Claude Code, OpenCode, OpenClaw, custom agents)","keyMetrics":[{"key":"model","label":"Best model","value":"42 AA","display":"42 AA"},{"key":"requests_per_month","label":"Requests/month","value":"172.8K R/mo","display":"172.8K R/mo"},{"key":"paid_value_ratio","label":"Value ratio","value":"0.05 value/$","display":"0.05 value/$"}],"url":"https://www.kimi.com/code"}},{"slug":"command-code-provider-api","description":"A direct provider API with included usage, discounted open models, and list-price access to premium frontier models.","priceFact":"$15/mo + PAYG","offer":{"slug":"command-code-provider-api","title":"Command Code Provider API","provider":"Command Code Provider API","planName":"paid","logo":"/logos/command-code-provider-api.png","pricingModel":"paid","accessBucket":"paid","freeLimitText":"Paid — $15/mo ($15 usage included, then PAYG at model rates, no markup). Raw OpenAI- + Anthropic-compatible API ('no lock-in — any client or agent') with GLM-5.2 + GLM-5.2 Fast (1M ctx), Kimi K2.7 Code + HighSpeed, DeepSeek V4 Pro (75% off), MiMo V2.5 (98% off), plus Claude/GPT/Gemini at list; $15 usage included, then pay-as-you-go with permanent discounts","keyMetrics":[{"key":"model","label":"Best model","value":"51.1 AA","display":"51.1 AA"}],"url":"https://commandcode.ai/pricing"}},{"slug":"deepseek-platform-api","description":"Low-cost metered DeepSeek API access for chat and reasoning workloads when pay-as-you-go pricing is the right fit.","priceFact":"$0.28/M out","offer":{"slug":"deepseek-platform-api","title":"DeepSeek Platform API","provider":"DeepSeek Platform API","planName":"paid","logo":"/logos/deepseek-platform-api.png","pricingModel":"paid","accessBucket":"paid","freeLimitText":"Paid — Pay-as-you-go, ~$0.28/M output (chat); R1 $0.55/$2.19. deepseek-chat (V3.2): $0.028 cache / $0.28 input miss / $0.42 output per 1M; deepseek-reasoner (R1): $0.14/$0.55/$2.19, cheapest top-tier reasoning","keyMetrics":[{"key":"model","label":"Best model","value":"24.7 AA","display":"24.7 AA"}],"url":"https://api-docs.deepseek.com/quick_start/pricing"}}],"specialSlot":{"slug":"nvidia","description":"NVIDIA NIM is a conditional developer-access offer: useful when its model catalog and rate limits match the prototype.","priceFact":"40 RPM","verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Live-verified: ongoing rate-limited trial (~40 RPM), no card, instant NGC key, 145+ models; commercial use NOT allowed on the free endpoint (NVIDIA API Trial ToS)."},"label":"40 RPM developer access","offer":{"slug":"nvidia","title":"NVIDIA NIM","provider":"NVIDIA NIM","planName":"freemium","logo":"/logos/nvidia.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free developer access: sign up and call many models through one endpoint; free use is governed by a 40 requests-per-minute rate limit (NVIDIA retired the old signup-credit system). Verify current rate limits and model list on the site.","keyMetrics":[{"key":"model","label":"Best model","value":"44.4 AA","display":"44.4 AA"},{"key":"requests_per_minute","label":"Requests/min","value":"40 R/min","display":"40 R/min"}],"url":"https://build.nvidia.com/"}}},"embeddings":{"collection":"infrastructure","category":"embeddings","updatedAt":"2026-07-10","intro":"Hosted vector and embedding services with free starter capacity for retrieval-augmented agent workflows.","picks":[{"slug":"pinecone","headline":"Mainstream managed vector search","description":"Pinecone is the hero embeddings pick because it is the most recognized managed vector database with a clear and generous free Starter plan. The free tier includes up to 2 GB storage, 2 million write units per month, 1 million read units per month, and up to 5 serverless indexes on AWS us-east-1, plus 5 million embedding tokens per month of inference and 1 GB of Assistant storage. That quota handles real retrieval prototypes and small production workloads before the paid Builder plan at $20 per month is needed.","whyChips":["2 GB storage","Serverless indexes","Inference included"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"pinecone","title":"Pinecone","provider":"Pinecone","planName":"freemium","logo":"/logos/pinecone-builder.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"The Starter plan is free with no credit card: up to 2 GB storage, 2M write units/mo, 1M read units/mo, and up to 5 serverless indexes on AWS us-east-1, plus 5M embedding tokens/mo of Inference and 1 GB of Assistant storage.","keyMetrics":[{"key":"free_tokens_per_month","label":"Monthly tokens","value":"5M T/mo","display":"5M T/mo"},{"key":"storage_gb","label":"Storage","value":"2 GB","display":"2 GB"}],"url":"https://www.pinecone.io/"}},{"slug":"weaviate-cloud-serverless","headline":"Free sandbox with lowest storage cost","description":"Weaviate Cloud Serverless is the side pick for teams that want a managed Weaviate instance without running the stack themselves. The free tier includes 100,000 objects and 1 GB of storage, which is enough to evaluate Weaviate's hybrid search and vector capabilities. Beyond the free tier, the per-dimension pricing at $0.095 per million dimensions is among the cheapest published rates for vector storage.","whyChips":["100k objects free","Hybrid search","$0.095/M dims"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"weaviate-cloud-serverless","title":"Weaviate Cloud","provider":"Weaviate Cloud","planName":"freemium","logo":"/logos/weaviate-cloud-serverless.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier (100K objects, 1 GB) to start; Serverless then ~$25/mo minimum, then $0.095 per 1M vector dimensions/month (storage) — among the cheapest per-dimension vector rates.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"1 GB","display":"1 GB"}],"url":"https://weaviate.io/pricing"}},{"slug":"zilliz-cloud","headline":"Managed Milvus with largest free quota","description":"Zilliz Cloud is the side pick for teams that want the largest free quota among the managed Milvus options. The free cluster includes 5 GB storage (approximately 1 million 768-dimensional vectors), 2.5 million vCUs per month, and up to 5 collections with no credit card required. As the creators of Milvus, the migration story and ecosystem support are clearer than for other managed vector stores.","whyChips":["5 GB storage","Milvus ecosystem","No card"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"zilliz-cloud","title":"Zilliz Cloud","provider":"Zilliz Cloud","planName":"freemium","logo":"/logos/zilliz-cloud-serverless.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"One Free cluster per organization with 5 GB storage (~1M 768-dim vectors), 2.5M vCUs/month, and up to 5 collections, no credit card required.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"5 GB","display":"5 GB"}],"url":"https://zilliz.com"}}],"hero":{"slug":"pinecone","headline":"Mainstream managed vector search","description":"Pinecone is the hero embeddings pick because it is the most recognized managed vector database with a clear and generous free Starter plan. The free tier includes up to 2 GB storage, 2 million write units per month, 1 million read units per month, and up to 5 serverless indexes on AWS us-east-1, plus 5 million embedding tokens per month of inference and 1 GB of Assistant storage. That quota handles real retrieval prototypes and small production workloads before the paid Builder plan at $20 per month is needed.","whyChips":["2 GB storage","Serverless indexes","Inference included"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"pinecone","title":"Pinecone","provider":"Pinecone","planName":"freemium","logo":"/logos/pinecone-builder.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"The Starter plan is free with no credit card: up to 2 GB storage, 2M write units/mo, 1M read units/mo, and up to 5 serverless indexes on AWS us-east-1, plus 5M embedding tokens/mo of Inference and 1 GB of Assistant storage.","keyMetrics":[{"key":"free_tokens_per_month","label":"Monthly tokens","value":"5M T/mo","display":"5M T/mo"},{"key":"storage_gb","label":"Storage","value":"2 GB","display":"2 GB"}],"url":"https://www.pinecone.io/"}},"nominations":[{"slug":"qdrant-cloud-free","label":"Small cluster","description":"Qdrant Cloud Free Tier provides a free-forever single-node cluster with 0.5 vCPU, 1 GB RAM, and 4 GB disk, plus free cloud inference on selected models. It is the nomination for developers who want a straightforward managed vector database with a generous free tier and clear upgrade path to dedicated infrastructure.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"qdrant-cloud-free","title":"Qdrant Cloud Free Tier","provider":"Qdrant Cloud Free Tier","planName":"free","logo":"/logos/qdrant-cloud-free.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free forever: single-node cluster with 0.5 vCPU, 1GB RAM, 4GB disk, plus free cloud inference with selected models.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"4 GB","display":"4 GB"}],"url":"https://qdrant.tech/pricing/"}},{"slug":"upstash-vector","label":"Serverless usage","description":"Upstash Vector is the nomination for serverless applications that want vector search with per-request billing and daily operation limits. The free tier includes 10,000 daily query and update operations with up to 1 GB of data, making it suitable for lightweight retrieval workloads where cost predictability matters.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"upstash-vector","title":"Upstash | Vector","provider":"Upstash","planName":"freemium","logo":"/logos/upstash.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free tier: 10K daily query/update operations, 1 GB max data/metadata size, 1,536 max dimensions, and 100 namespaces.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"1 GB","display":"1 GB"}],"url":"https://upstash.com/docs/vector"}}],"paidPicks":[]},"databases":{"collection":"infrastructure","category":"databases","updatedAt":"2026-07-10","intro":"Managed databases with durable free or conditional capacity.","picks":[{"slug":"supabase","headline":"The full backend Postgres default","description":"Supabase is the hero database pick because it pairs a managed Postgres instance with auth, file storage, real-time subscriptions, and edge functions in one project. The free tier gives a workable project for prototypes and small products without a credit card. It is the lowest-friction path from zero to a complete hosted backend.","whyChips":["No card","Full backend","Auth + storage"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"supabase","title":"Supabase","provider":"Supabase","planName":"freemium","logo":"/logos/supabase-pro.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free project tier; storage/database quotas depend on current plan.","keyMetrics":[{"key":"bandwidth_gb","label":"Egress","value":"5 GB/mo","display":"5 GB/mo"},{"key":"max_databases","label":"Databases","value":"2 DBs","display":"2 DBs"}],"url":"https://supabase.com/"}},{"slug":"neon","headline":"Non-branchable serverless Postgres","description":"Neon is a strong side pick because it separates compute and storage, giving you serverless Postgres that scales to zero without cold-start latency on every connection. Branching databases for staging and preview environments is a distinctive feature that fits agent development workflows where each agent run needs an isolated data view.","whyChips":["Scales to zero","Branching","Serverless Postgres"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"neon","title":"Neon","provider":"Neon","planName":"free_forever","logo":"/logos/neon-launch.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Free tier available; compute/storage limits depend on plan.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"0.5 GB","display":"0.5 GB"},{"key":"bandwidth_gb","label":"Egress","value":"5 GB/mo","display":"5 GB/mo"},{"key":"compute_hours","label":"Compute","value":"100 hrs/mo","display":"100 hrs/mo"}],"url":"https://neon.com/"}},{"slug":"turso","headline":"SQLite close to the edge","description":"Turso is the side pick for applications that benefit from embedded SQLite semantics replicated to the edge. The free starter tier covers small production experiments and agent state use cases. The libSQL compatibility also provides a familiar interface for Postgres migrants who want multi-tenant isolation without the overhead of a full Postgres cluster.","whyChips":["Edge SQLite","libSQL","Free starter"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"turso","title":"Turso","provider":"Turso","planName":"freemium","logo":"/logos/turso-developer.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free starter tier available; rows/storage/groups vary by plan.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"5 GB","display":"5 GB"},{"key":"row_writes_m","label":"Writes","value":"10 M rows/mo","display":"10 M rows/mo"},{"key":"bandwidth_gb","label":"Egress","value":"3 GB/mo","display":"3 GB/mo"}],"url":"https://turso.tech/"}}],"hero":{"slug":"supabase","headline":"The full backend Postgres default","description":"Supabase is the hero database pick because it pairs a managed Postgres instance with auth, file storage, real-time subscriptions, and edge functions in one project. The free tier gives a workable project for prototypes and small products without a credit card. It is the lowest-friction path from zero to a complete hosted backend.","whyChips":["No card","Full backend","Auth + storage"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"supabase","title":"Supabase","provider":"Supabase","planName":"freemium","logo":"/logos/supabase-pro.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free project tier; storage/database quotas depend on current plan.","keyMetrics":[{"key":"bandwidth_gb","label":"Egress","value":"5 GB/mo","display":"5 GB/mo"},{"key":"max_databases","label":"Databases","value":"2 DBs","display":"2 DBs"}],"url":"https://supabase.com/"}},"nominations":[{"slug":"cockroachdb-basic","label":"Scale-to-zero","description":"CockroachDB Basic includes 50 million RUs and 10 GiB storage free per month on the Basic plan, scales to zero between uses, and requires no credit card. It is a practical way to get a globally resilient, Postgres-compatible SQL database without managing infrastructure or committing to a paid plan.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"cockroachdb-basic","title":"CockroachDB Cloud","provider":"CockroachDB Cloud","planName":"freemium","logo":"/logos/cockroachdb-basic.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"50 million Request Units (RUs) and 10 GiB storage free per month on the Basic plan; scales to zero; no credit card required for Basic. ($400 in trial credits also available separately.)","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"max_databases","label":"Databases","value":"1 DBs","display":"1 DBs"}],"url":"https://www.cockroachlabs.com/pricing/"}},{"slug":"prisma-postgres","label":"Postgres in one click","description":"Prisma Postgres is directly integrated with the Prisma ORM, making it a natural choice for teams already using Prisma for schema management. The free tier includes 100,000 operations per month, 500 MB storage, and 50 databases with no card required, which is enough to evaluate a small project before committing to a paid plan.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"prisma-postgres","title":"Prisma Postgres","provider":"Prisma Postgres","planName":"freemium","logo":"/logos/prisma-postgres-starter.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free: 100,000 operations/month included, 500 MB storage, 50 databases, no credit card required.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"0.5 GB","display":"0.5 GB"},{"key":"bandwidth_gb","label":"Egress","value":"∞","display":"∞"},{"key":"max_databases","label":"Databases","value":"50 DBs","display":"50 DBs"}],"url":"https://www.prisma.io/pricing"}}],"paidPicks":[]},"app-platforms":{"collection":"infrastructure","category":"app-platforms","updatedAt":"2026-07-10","intro":"Full-stack and backend hosting with credible free deploy paths.","picks":[{"slug":"convex-free","headline":"The reactive backend-in-a-box","description":"Convex is the hero app-platform pick because it combines a reactive Postgres-like database, serverless functions, scheduled jobs, auth, file storage, and vector search into one coherent platform. The free and Starter tiers cover up to 40 deployments and 1 million function calls per month, which is enough to build and ship a real agent app without stitching together separate services. The reactive programming model reduces the boilerplate of keeping agent state synchronized across clients.","whyChips":["Reactive database","Functions + cron","1M calls/mo"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"convex-free","title":"Convex","provider":"Convex","planName":"free","logo":"/logos/convex-free.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free & Starter: reactive database, file/text/vector search, crons, auth, Node.js actions, up to 40 deployments (1M function calls/month).","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"1 GB/mo","display":"1 GB/mo"}],"url":"https://convex.dev/pricing"}},{"slug":"koyeb-free","headline":"Simple container platform plus Postgres","description":"Koyeb is the side pick for teams that want a simple container platform with a managed Postgres option and a path to serverless GPUs without Kubernetes complexity. The free tier gives one web service with 512 MB RAM, 0.1 vCPU, 2 GB SSD, plus a free PostgreSQL database with 1 GB storage and 100 GB monthly outbound bandwidth. Per-second billing means you only pay for what you use once production traffic starts.","whyChips":["Containers + Postgres","GPU path","Free tier"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"koyeb-free","title":"Koyeb","provider":"Koyeb","planName":"free","logo":"/logos/koyeb.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"One free web Service (512 MB RAM, 0.1 vCPU, 2 GB SSD) in Frankfurt, Washington D.C., or Singapore, plus one free PostgreSQL database (1 GB storage, 5 hours active time), and 100 GB free outbound bandwidth/month. Requires a credit card (a $29 pre-authorization hold is placed during verification).","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"100 GB/mo","display":"100 GB/mo"},{"key":"build_minutes_mo","label":"Build minutes","value":"∞","display":"∞"},{"key":"custom_domains","label":"Custom domains","value":"5 domains","display":"5 domains"}],"url":"https://www.koyeb.com/docs/faqs/pricing"}},{"slug":"huggingface-spaces","headline":"Best place to publish an AI demo","description":"Hugging Face Spaces is the side pick for agent apps that need to ship an AI-facing demo to collaborators or users quickly. Free CPU Spaces exist and the community discovery on the platform gives prototypes visibility without any hosting setup. The trade-off is that free CPU Spaces sleep and restart, so they are not reliable for always-on bots or real-time agents — but for shareable demos, evaluations, and prototypes, the path from zero to a live demo is the shortest available.","whyChips":["AI demos","Community discovery","Free CPU"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"huggingface-spaces","title":"Hugging Face Spaces","provider":"Hugging Face Spaces","planName":"freemium","logo":"/logos/huggingface.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free CPU Spaces exist; sleeping/restarts make realtime bots unreliable.","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"∞","display":"∞"},{"key":"build_minutes_mo","label":"Build minutes","value":"∞","display":"∞"}],"url":"https://huggingface.co/spaces"}}],"hero":{"slug":"convex-free","headline":"The reactive backend-in-a-box","description":"Convex is the hero app-platform pick because it combines a reactive Postgres-like database, serverless functions, scheduled jobs, auth, file storage, and vector search into one coherent platform. The free and Starter tiers cover up to 40 deployments and 1 million function calls per month, which is enough to build and ship a real agent app without stitching together separate services. The reactive programming model reduces the boilerplate of keeping agent state synchronized across clients.","whyChips":["Reactive database","Functions + cron","1M calls/mo"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"convex-free","title":"Convex","provider":"Convex","planName":"free","logo":"/logos/convex-free.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free & Starter: reactive database, file/text/vector search, crons, auth, Node.js actions, up to 40 deployments (1M function calls/month).","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"1 GB/mo","display":"1 GB/mo"}],"url":"https://convex.dev/pricing"}},"nominations":[{"slug":"leapcell","label":"Always-on","description":"Leapcell's Hobby plan covers up to 20 projects with 3 vCPU and 4 GB RAM each, 100,000 serverless invocations, 3 GB-hours compute, 1 free PostgreSQL database, 100K Redis commands per month, and 2 GB egress per month — all at no cost with no credit card required. The offering is distinctive because it includes a real always-on Postgres alongside serverless functions, which is rare at the free tier.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"leapcell","title":"Leapcell","provider":"Leapcell","planName":"freemium","logo":"/logos/leapcell.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Hobby plan ($0): up to 20 projects (3 vCPU / 4 GB RAM each), 100,000 serverless invocations and 3 GB-hours compute per month, 1 free PostgreSQL database, 100K Redis commands/mo, 2 GB egress/mo, 60 build min/mo, plus custom domains with free TLS and global CDN.","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"2 GB/mo","display":"2 GB/mo"},{"key":"build_minutes_mo","label":"Build minutes","value":"60 min/mo","display":"60 min/mo"}],"url":"https://leapcell.io"}},{"slug":"nhost-free","label":"Full backend","description":"Nhost is the nomination for teams that want a GraphQL-first managed backend with Postgres, Hasura, auth, storage, and serverless functions already wired together. The free Starter plan includes 1 GB database, 1 GB storage, 5 GB egress, and unlimited users — enough for a small production experiment or a weekend project that does not want to configure each service individually.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"nhost-free","title":"Nhost","provider":"Nhost","planName":"free","logo":"/logos/nhost-free.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Starter $0/month: 1 GB database, 1 GB storage, 5 GB egress, functions, realtime APIs, unlimited users, OAuth, 1 project (paused after 1 week idle).","keyMetrics":[{"key":"bandwidth_gb_mo","label":"Bandwidth","value":"5 GB/mo","display":"5 GB/mo"},{"key":"function_invocations_mo","label":"Function calls","value":"∞","display":"∞"}],"url":"https://nhost.io/pricing"}}],"paidPicks":[]},"agent-deployment":{"collection":"infrastructure","category":"agent-deployment","updatedAt":"2026-07-12","intro":"Hosted agent runtimes with useful free or starter capacity.","picks":[{"slug":"cloudflare-agents","headline":"Best free stateful agent runtime","description":"Cloudflare Agents is the strongest default pick for a hosted agent runtime because it combines Workers, Durable Objects, and a large no-card free allocation. The free plan gives enough daily request and stateful compute capacity for prototypes, internal tools, and small public agents. It is also close to the edge platform many teams already use, so the path from experiment to production is short.","whyChips":["No card","Stateful runtime","Edge deploys"],"verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Workers Free 100k req/day + DO 100k req/day / 13,000 GB-s/day confirmed live on the Durable Objects pricing page; no card."},"offer":{"slug":"cloudflare-agents","title":"Cloudflare Agents","provider":"Cloudflare Agents","planName":"free","logo":"/logos/cloudflare.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Workers Free plan: 100,000 requests/day, 10 ms CPU/invocation, unlimited bandwidth. Durable Objects (which back each agent's state) are included free: 100,000 requests/day and 13,000 GB-s/day of compute. No credit card required. Workers AI inference has a separate free allocation.","keyMetrics":[{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"4 signals","display":"4 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://agents.cloudflare.com"}},{"slug":"lyzr-agent-studio","headline":"Generous no-code agent studio","description":"Lyzr is a useful pick when the goal is to build and test multiple hosted agents without starting from infrastructure code. The community tier lists meaningful monthly credits, agent slots, and knowledge-base capacity. It is a good editorial counterpoint to code-first runtimes.","whyChips":["Hosted studio","Monthly credits"],"verified":{"verifiedAt":"2026-07-12","confidence":"medium","note":"Community Plan ($0, 10 agents, 500 credits/mo) confirmed in docs.lyzr.ai/introduction/plans, but the main pricing page now leads with per-run pricing only."},"offer":{"slug":"lyzr-agent-studio","title":"Lyzr Agent Studio","provider":"Lyzr Agent Studio","planName":"free","logo":"/logos/lyzr-agent-studio.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Ongoing free tier on Lyzr Agent Studio (hosted Lyzr Cloud). Two figures appear in Lyzr's own docs: the Plans page lists a \"Community Plan\" at $0/month with 500 credits/month (1 credit = $1), build up to 10 agents, 5 knowledge bases, 100 MB RAG storage, 5 tools, 1 builder license, 7-day logs; the Cre","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"1 signals","display":"1 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://www.lyzr.ai"}},{"slug":"openai-agentkit-agent-builder-chatkit-agents-sdk","headline":"Best builder experience","description":"OpenAI AgentKit is notable for Agent Builder, ChatKit, and the Agents SDK living in one workflow. The builder surface is free to use, while runtime cost follows model usage. Pick it when design, handoff, and SDK alignment matter more than raw free compute.","whyChips":["Visual builder","SDK aligned"],"verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Agent Builder/ChatKit/Agents SDK included with standard API model pricing (openai.com AgentKit announcement); runtime = paid tokens."},"offer":{"slug":"openai-agentkit-agent-builder-chatkit-agents-sdk","title":"OpenAI AgentKit","provider":"OpenAI AgentKit","planName":"limited","logo":"/logos/openai-agentkit-agent-builder-chatkit-agents-sdk.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Agent Builder canvas is free to use; you pay only for model/token usage when agents run","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"6 signals","display":"6 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://openai.com/index/introducing-agentkit/"}}],"hero":{"slug":"cloudflare-agents","headline":"Best free stateful agent runtime","description":"Cloudflare Agents is the strongest default pick for a hosted agent runtime because it combines Workers, Durable Objects, and a large no-card free allocation. The free plan gives enough daily request and stateful compute capacity for prototypes, internal tools, and small public agents. It is also close to the edge platform many teams already use, so the path from experiment to production is short.","whyChips":["No card","Stateful runtime","Edge deploys"],"verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Workers Free 100k req/day + DO 100k req/day / 13,000 GB-s/day confirmed live on the Durable Objects pricing page; no card."},"offer":{"slug":"cloudflare-agents","title":"Cloudflare Agents","provider":"Cloudflare Agents","planName":"free","logo":"/logos/cloudflare.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Workers Free plan: 100,000 requests/day, 10 ms CPU/invocation, unlimited bandwidth. Durable Objects (which back each agent's state) are included free: 100,000 requests/day and 13,000 GB-s/day of compute. No credit card required. Workers AI inference has a separate free allocation.","keyMetrics":[{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"4 signals","display":"4 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://agents.cloudflare.com"}},"nominations":[{"slug":"modal","label":"Compute credits","description":"Starter credits and strong serverless primitives make Modal a practical choice for agent jobs that need containers or GPUs.","verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Starter $0/mo with $30/month recurring free credits, 100 containers + 10 GPU concurrency confirmed live on modal.com/pricing."},"icon":"medal","offer":{"slug":"modal","title":"Modal","provider":"Modal","planName":"paid","logo":"/logos/modal.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Paid — $0/mo (Starter) + usage; CPU $0.0000131/core-sec. Starter free with $30/mo free compute credits, 100 containers + 10 GPU concurrency, 5 crons, 8 webhooks. Sandbox CPU $0.00003942/core-sec, memory $0.00000672/GiB-sec, volumes $0.09/GiB-mo (1 TiB free). Team $250/mo + comput","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"1 signals","display":"1 signals"},{"key":"agent_production_controls","label":"Production controls","value":"2 signals","display":"2 signals"}],"url":"https://modal.com/pricing"}},{"slug":"beam-cloud","label":"GPU starter","description":"Beam Cloud stands out for a developer tier with recurring credits and simple container deployment for AI workloads.","verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Developer tier $0/mo with $30 monthly credits and 5 GPU container concurrency confirmed live on beam.cloud/pricing."},"icon":"medal","offer":{"slug":"beam-cloud","title":"Beam Cloud","provider":"Beam Cloud","planName":"paid","logo":"/logos/beam-cloud.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Paid — $0/mo (Developer) + usage; CPU $0.0000528/core-sec. Developer free with $30/mo credits, unlimited apps, 5 GPU + 30 CPU containers. CPU $0.0000528/core-sec, RAM $0.0000056/GB-sec, RTX 4090 $0.000192/sec; on-demand H100 $1.74/hr, A100 80GB $1.30/hr; storage free; cold-start ","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"1 signals","display":"1 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://www.beam.cloud/pricing"}},{"slug":"arcade-dev","label":"Tool execution","description":"Arcade.dev is worth tracking for hosted tool execution, user authorization, and MCP-oriented agent integrations.","verified":{"verifiedAt":"2026-07-12","confidence":"high","note":"Hobby FREE: 1,000 standard + 50 pro tool executions, 100 auth challenges, 1 hosted MCP server 24/7 confirmed live on arcade.dev/pricing."},"icon":"medal","offer":{"slug":"arcade-dev","title":"Arcade.dev","provider":"Arcade.dev","planName":"free","logo":"/logos/arcade-dev.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Hobby plan (FREE, $0 platform fee, ongoing): 1,000 standard tool executions, 50 pro tool executions, 100 user authorization challenges, and 1 Arcade-hosted MCP server running 24/7 included. Shared tenant. Community support via GitHub Discussions. Rate limit 1,000/min.","keyMetrics":[{"key":"agent_execution_quota","label":"Execution quota","value":"1 runs/month","display":"1 runs/month"},{"key":"agent_tool_model_connectivity","label":"Tool/model connectivity","value":"4 signals","display":"4 signals"},{"key":"agent_production_controls","label":"Production controls","value":"1 signals","display":"1 signals"}],"url":"https://www.arcade.dev/pricing"}}],"paidPicks":[]},"serverless-functions":{"collection":"infrastructure","category":"serverless-functions","updatedAt":"2026-07-10","intro":"Serverless execution with practical free developer capacity at the edge and beyond.","picks":[{"slug":"cloudflare-workers","headline":"The edge function default","description":"Cloudflare Workers is the hero serverless pick because it runs JavaScript and TypeScript at the edge with a genuinely useful free tier and a pricing model that stays simple as usage grows. The runtime is fast, the cold-start is near-zero, and Workers are tightly integrated with Cloudflare's KV, D1, R2, and Queues — which means agent logic can stay within one provider's ecosystem without sacrificing performance. The 100,000 requests per day free allowance covers meaningful prototype and hobby workloads.","whyChips":["100k req/day","Edge runtime","KV + D1 + R2"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudflare-workers","title":"Cloudflare Workers","provider":"Cloudflare Workers","planName":"freemium","logo":"/logos/cloudflare-workers-paid.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier historically centered around daily request quotas; verify current limits before launch.","keyMetrics":[{"key":"max_cpu_per_invocation","label":"CPU/call","value":"10 ms","display":"10 ms"},{"key":"bandwidth_per_month","label":"Bandwidth","value":"∞","display":"∞"}],"url":"https://developers.cloudflare.com/workers/"}},{"slug":"val-town","headline":"Fastest script-to-endpoint path","description":"Val Town earns its pick as the fastest path from a script to a hosted endpoint. The free plan gives 100,000 runs per day at 15-minute intervals, unlimited public vals, and MCP integration — enough to build cron jobs, webhooks, and small AI-connected endpoints without any deployment pipeline. The wall-clock limit and public-only constraint on the free plan are honest trade-offs that match the simplicity of the use case.","whyChips":["100k runs/day","No deploy","MCP integration"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"val-town","title":"Val Town","provider":"Val Town","planName":"freemium","logo":"/logos/val-town.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan: 100,000 runs/day, unlimited public vals, 15-minute cron intervals, 1-minute wall-clock time per run, 3-day log retention, MCP integration. (Free tier allows 0 private vals.)","keyMetrics":[{"key":"compute_gb_seconds","label":"Compute","value":"∞","display":"∞"},{"key":"max_cpu_per_invocation","label":"CPU/call","value":"60K ms","display":"60K ms"},{"key":"bandwidth_per_month","label":"Bandwidth","value":"∞","display":"∞"}],"url":"https://www.val.town/pricing"}},{"slug":"deno-deploy","headline":"Modern TypeScript at the edge","description":"Deno Deploy is the side pick for teams that want a first-class TypeScript runtime with built-in KV and volume storage at the edge. The free plan includes 1 million inbound requests per month, 20 GB egress, 1 GiB volume storage, and Deno KV with 450k read and 300k write units per month — a generous allowance for a modern edge-native stack that avoids Node.js compatibility overhead.","whyChips":["TypeScript edge","KV included","1M req + 20GB egress"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"deno-deploy","title":"Deno Deploy","provider":"Deno Deploy","planName":"freemium","logo":"/logos/deno-deploy-pro.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan: 1M inbound requests/month, 20 GB egress, 1 GiB volume storage, 1 GiB Deno KV storage, 450k KV read units and 300k KV write units per month, 50 custom domains.","keyMetrics":[{"key":"bandwidth_per_month","label":"Bandwidth","value":"20 GB/mo","display":"20 GB/mo"}],"url":"https://deno.com/deploy/pricing"}}],"hero":{"slug":"cloudflare-workers","headline":"The edge function default","description":"Cloudflare Workers is the hero serverless pick because it runs JavaScript and TypeScript at the edge with a genuinely useful free tier and a pricing model that stays simple as usage grows. The runtime is fast, the cold-start is near-zero, and Workers are tightly integrated with Cloudflare's KV, D1, R2, and Queues — which means agent logic can stay within one provider's ecosystem without sacrificing performance. The 100,000 requests per day free allowance covers meaningful prototype and hobby workloads.","whyChips":["100k req/day","Edge runtime","KV + D1 + R2"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudflare-workers","title":"Cloudflare Workers","provider":"Cloudflare Workers","planName":"freemium","logo":"/logos/cloudflare-workers-paid.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier historically centered around daily request quotas; verify current limits before launch.","keyMetrics":[{"key":"max_cpu_per_invocation","label":"CPU/call","value":"10 ms","display":"10 ms"},{"key":"bandwidth_per_month","label":"Bandwidth","value":"∞","display":"∞"}],"url":"https://developers.cloudflare.com/workers/"}},"nominations":[{"slug":"modal","label":"GPU compute","description":"Modal is the nomination for workloads that need GPU acceleration, Python containers, or recurring monthly compute credits rather than per-invocation pricing. The Starter plan includes $30 in free compute credits each month, which covers scheduled GPU inference jobs, batch processing, and sandboxed AI agent loops before the paid threshold arrives.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"modal","title":"Modal","provider":"Modal","planName":"freemium","logo":"/logos/modal.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Starter plan ($0): $30/month in free compute credits (recurring), 3 workspace seats, 100 containers + 10 GPU concurrency, 5 deployed crons, 8 webhook endpoints, 200 deployed apps, 1 TiB/month free volume storage.","keyMetrics":[{"key":"bandwidth_per_month","label":"Bandwidth","value":"∞","display":"∞"}],"url":"https://modal.com/pricing"}},{"slug":"upstash-workflow","label":"Durable workflow steps","description":"Upstash Workflow is the nomination for serverless functions that are really sequences of durable steps — read a document, call an LLM, write a result, send a notification. The free tier includes 1,000 steps per day and 15-minute maximum HTTP response duration, which covers lightweight agent loops without managing a queue service separately.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"upstash-workflow","title":"Upstash | Workflow","provider":"Upstash","planName":"freemium","logo":"/logos/upstash.svg","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free tier: 1,000 steps/day, 1,000 steps per workflow run, 10 concurrent steps, 1 MB max message size, 7 day max sleep, and 15 min max HTTP response duration.","keyMetrics":[{"key":"bandwidth_per_month","label":"Bandwidth","value":"50 GB/mo","display":"50 GB/mo"}],"url":"https://upstash.com/docs/workflow"}}],"paidPicks":[]},"schedulers":{"collection":"infrastructure","category":"schedulers","updatedAt":"2026-07-10","intro":"Cron and background work services for recurring jobs, checks, queues, and workflows.","picks":[{"slug":"val-town-cron","headline":"The fastest cron for small scripts","description":"Val Town Cron is the hero scheduler pick because it turns a small scheduled script into a hosted recurring job with almost no setup. The free plan supports public vals and 15-minute cron intervals, 100,000 runs per day, and MCP integration — enough to cover most lightweight agent scheduling needs without configuring a pipeline or managing infrastructure. The wall-clock limit on the free plan keeps it honest for its weight class.","whyChips":["100k runs/day","15-min intervals","MCP integration"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"val-town-cron","title":"Val Town Cron","provider":"Val Town Cron","planName":"freemium","logo":"/logos/val-town.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan supports public vals and cron intervals; private vals and higher frequency require checking the current plan.","keyMetrics":[{"key":"cron_jobs","label":"Cron jobs","value":"2.5K jobs","display":"2.5K jobs"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"6 signals","display":"6 signals"},{"key":"min_interval","label":"Min interval","value":"15 min","display":"15 min"}],"url":"https://docs.val.town/vals/cron/"}},{"slug":"healthchecks-io","headline":"Dead-man switch for cron","description":"Healthchecks.io is the side pick for agent developers who want a simple dead-man switch: a service that alerts when a scheduled job fails to fire. The free plan includes 20 free checks, and the paid plan at roughly $16 per month gives 100 checks with longer log retention — making it a cheap insurance policy for agent workflows where missed runs have real consequences.","whyChips":["20 free checks","Alert on miss","Log retention"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"healthchecks-io","title":"Healthchecks.io","provider":"Healthchecks.io","planName":"paid","logo":"/logos/healthchecks-io.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Paid — $16/mo (Business, $192/yr). 100 checks (crons) at the Business plan ($192/yr, $16/mo) with longer log retention; 20 free checks; free Business plan for OSS/nonprofits. Alerts when jobs don't run on time.","keyMetrics":[{"key":"scheduler_run_quota","label":"Run quota","value":"1 runs/month","display":"1 runs/month"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"2 signals","display":"2 signals"}],"url":"https://healthchecks.io/pricing/"}},{"slug":"inngest","headline":"Durable workflows, not just cron","description":"Inngest earns its pick as the durable workflow scheduler for agent pipelines that go beyond simple cron: retries, branching, fan-out, and event-driven steps are first-class primitives rather than bolt-ons. The free Hobby plan includes 50,000 executions per month, 500k events per month, and unlimited serverless cron functions, with no card required. That quota handles meaningful agent workflow prototypes before scaling to a paid plan.","whyChips":["50k executions/mo","Durable retries","Unlimited cron"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"inngest","title":"Inngest","provider":"Inngest","planName":"freemium","logo":"/logos/inngest.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Hobby plan ($0/mo, no card): 50,000 executions/month, 500k events/month, 5 concurrent steps, 100k queue depth, unlimited serverless cron functions, and 24h trace/log retention.","keyMetrics":[{"key":"cron_jobs","label":"Cron jobs","value":"∞","display":"∞"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"12 signals","display":"12 signals"},{"key":"min_interval","label":"Min interval","value":"1 min","display":"1 min"}],"url":"https://www.inngest.com"}}],"hero":{"slug":"val-town-cron","headline":"The fastest cron for small scripts","description":"Val Town Cron is the hero scheduler pick because it turns a small scheduled script into a hosted recurring job with almost no setup. The free plan supports public vals and 15-minute cron intervals, 100,000 runs per day, and MCP integration — enough to cover most lightweight agent scheduling needs without configuring a pipeline or managing infrastructure. The wall-clock limit on the free plan keeps it honest for its weight class.","whyChips":["100k runs/day","15-min intervals","MCP integration"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"val-town-cron","title":"Val Town Cron","provider":"Val Town Cron","planName":"freemium","logo":"/logos/val-town.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free plan supports public vals and cron intervals; private vals and higher frequency require checking the current plan.","keyMetrics":[{"key":"cron_jobs","label":"Cron jobs","value":"2.5K jobs","display":"2.5K jobs"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"6 signals","display":"6 signals"},{"key":"min_interval","label":"Min interval","value":"15 min","display":"15 min"}],"url":"https://docs.val.town/vals/cron/"}},"nominations":[{"slug":"windmill","label":"Unlimited cron","description":"Windmill's Community Edition is self-hosted and completely free: unlimited cron schedules and executions across 3 workspaces with up to 50 users. It is the nomination for teams that want an open-source, code-first scheduler with a visual builder and execution history, and are comfortable operating the infrastructure themselves.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"windmill","title":"Windmill","provider":"Windmill","planName":"open_source_selfhost","logo":"/logos/paid-windmill.png","pricingModel":"open_source_selfhost","accessBucket":"conditional","freeLimitText":"Self-hosted Community Edition ($0): unlimited cron schedules and executions across 3 workspaces and up to 50 users (10 with SSO), with unlimited scripts/flows/apps and run-detail retention up to 30 days.","keyMetrics":[{"key":"scheduler_run_quota","label":"Run quota","value":"1 runs/month","display":"1 runs/month"},{"key":"cron_jobs","label":"Cron jobs","value":"∞","display":"∞"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"4 signals","display":"4 signals"}],"url":"https://www.windmill.dev"}},{"slug":"vercel-queues","label":"Most runs","description":"Vercel Queues includes the first 1 million Queue API operations per month free in the Hobby tier — a higher operational quota than most competing services at the free level. It is the nomination for agent developers already on Vercel who need to offload expensive work like emails, PDFs, and external API calls from synchronous request paths.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"vercel-queues","title":"Vercel Queues","provider":"Vercel Queues","planName":"freemium","logo":"/logos/vercel.svg","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Hobby includes the first 1,000,000 Queue API operations per month free; beyond that, regional on-demand pricing starting at $0.60 per 1M operations.","keyMetrics":[{"key":"executions","label":"API operations","value":"1M ops/mo","display":"1M ops/mo"},{"key":"scheduler_workflow_runtime","label":"Workflow/runtime","value":"9 signals","display":"9 signals"},{"key":"cron_jobs","label":"Topics","value":"∞","display":"∞"}],"url":"https://vercel.com/docs/queues"}}],"paidPicks":[]},"storage":{"collection":"infrastructure","category":"storage","updatedAt":"2026-07-10","intro":"Persistent, durable, bindable storage where free quota and egress economics matter.","picks":[{"slug":"cloudflare-r2-standard","headline":"The zero-egress object storage default","description":"Cloudflare R2 is the hero storage pick because it removes the cost developers fear most: egress to the public internet is free at every tier. The free tier includes 10 GB-months of storage, 1 million Class A operations and 10 million Class B operations per month. The S3-compatible API means existing tooling migrates without rewrite, and pairing with Cloudflare Workers keeps agent artifacts close to the compute that processes them.","whyChips":["$0 egress","10 GB storage","1M ops/mo write ops"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudflare-r2-standard","title":"Cloudflare R2","provider":"Cloudflare R2","planName":"freemium","logo":"/logos/cloudflare-r2-infrequent-access.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free 10 GB stored + 1M Class A ops + 10M Class B ops each month, with ZERO egress fees and no minimum retention; beyond the free tier storage is $0.015/GB-month.","keyMetrics":[{"key":"egress_gb","label":"Egress","value":"$0 egress","display":"$0 egress"},{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"class_a_ops","label":"Write ops","value":"1 M ops/mo","display":"1 M ops/mo"}],"url":"https://developers.cloudflare.com/r2/pricing/"}},{"slug":"backblaze-b2","headline":"Single cheap object storage","description":"Backblaze B2 is the plain-spoken storage pick: 10 GB free from the start, egress credits to reduce bills, and free egress up to 3x of stored data under the free-for-dev program. S3-compatible API, no credit card to create an account, and a straightforward pricing model make it a reliable baseline for backups, media assets, and agent output storage.","whyChips":["10 GB storage","3x free egress","S3-compatible"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"backblaze-b2","title":"Backblaze B2","provider":"Backblaze B2","planName":"freemium","logo":"/logos/backblaze-b2.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free 10 GB object storage for unlimited time (per free-for-dev); free egress up to 3x stored data; no credit card to create an account.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"egress_gb","label":"Free egress","value":"30 GB/mo","display":"30 GB/mo"},{"key":"class_a_ops","label":"Write ops","value":"∞","display":"∞"}],"url":"https://www.backblaze.com/cloud-storage"}},{"slug":"storj","headline":"Distributed storage with a no-card path","description":"Storj is the storage pick for applications where decentralized durability and S3-compatible storage are part of the value proposition. Accounts without a payment method get 25 GB of storage and 25 GB of egress per month with no monthly fee and no credit card required. The trade-off is that the free tier uses shared infrastructure, so it suits non-critical assets and archival more than hot production data.","whyChips":["25 GB storage","25 GB egress","No card"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"storj","title":"Storj","provider":"Storj","planName":"freemium","logo":"/logos/paid-storj.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Accounts without a payment method get 25 GB of storage and 25 GB of egress per month with no monthly fee and no credit card required.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"25 GB","display":"25 GB"},{"key":"egress_gb","label":"Free egress","value":"25 GB/mo","display":"25 GB/mo"}],"url":"https://www.storj.io"}}],"hero":{"slug":"cloudflare-r2-standard","headline":"The zero-egress object storage default","description":"Cloudflare R2 is the hero storage pick because it removes the cost developers fear most: egress to the public internet is free at every tier. The free tier includes 10 GB-months of storage, 1 million Class A operations and 10 million Class B operations per month. The S3-compatible API means existing tooling migrates without rewrite, and pairing with Cloudflare Workers keeps agent artifacts close to the compute that processes them.","whyChips":["$0 egress","10 GB storage","1M ops/mo write ops"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudflare-r2-standard","title":"Cloudflare R2","provider":"Cloudflare R2","planName":"freemium","logo":"/logos/cloudflare-r2-infrequent-access.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free 10 GB stored + 1M Class A ops + 10M Class B ops each month, with ZERO egress fees and no minimum retention; beyond the free tier storage is $0.015/GB-month.","keyMetrics":[{"key":"egress_gb","label":"Egress","value":"$0 egress","display":"$0 egress"},{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"class_a_ops","label":"Write ops","value":"1 M ops/mo","display":"1 M ops/mo"}],"url":"https://developers.cloudflare.com/r2/pricing/"}},"nominations":[{"slug":"tigris","label":"Zero egress","description":"Tigris gives S3-standard storage with 5 GB per month free, 10,000 Class A and 100,000 Class B requests free, and zero egress fees across all tiers. It is a clean pick when you want S3 compatibility without bandwidth bills, especially for data-intensive agent workloads that read more than they write.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"tigris","title":"Tigris Data","provider":"Tigris Data","planName":"freemium","logo":"/logos/tigris.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"5GB of Standard storage/month free, 10,000 Class A requests/month free, 100,000 Class B requests/month free, and zero egress fees on all tiers.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"5 GB","display":"5 GB"},{"key":"egress_gb","label":"Free egress","value":"∞","display":"∞"},{"key":"class_b_ops","label":"Read ops","value":"0.1 M ops/mo","display":"0.1 M ops/mo"}],"url":"https://www.tigrisdata.com/pricing/"}},{"slug":"telnyx-cloud-storage","label":"Most free GB","description":"Telnyx Cloud Storage includes 10 GiB of storage, 1 million state-change write operations, and 10 million state-read operations per month, all with no egress fees. The quota is unusually generous among the conditional-free tiers, making it a practical choice for agent apps that accumulate media or document assets.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"telnyx-cloud-storage","title":"Telnyx Cloud Storage","provider":"Telnyx Cloud Storage","planName":"freemium","logo":"/logos/paid-telnyx-cloud-storage.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"10 GiB of storage, 1 million state-change (write) operations, and 10 million state-read operations per month, with no egress fees.","keyMetrics":[{"key":"storage_gb","label":"Storage","value":"10 GB","display":"10 GB"},{"key":"egress_gb","label":"Free egress","value":"∞","display":"∞"},{"key":"class_a_ops","label":"Write ops","value":"1 M ops/mo","display":"1 M ops/mo"}],"url":"https://telnyx.com/products/cloud-storage"}}],"paidPicks":[]},"workspaces":{"collection":"infrastructure","category":"workspaces","updatedAt":"2026-07-10","intro":"Team workspaces, docs, and knowledge tools with free plans for agent project management.","picks":[{"slug":"grist","headline":"The relational database workspace","description":"Grist is the hero workspace pick because it is a spreadsheet-database hybrid that non-engineers can use fluently and engineers can extend with Python. The Cloud Free plan allows unlimited members and 2 guests per doc with 1 GB attachment storage per doc, and self-hosting is free for individuals and orgs with under $1M annual income. That combination makes it a durable, self-hostable foundation for agent-run projects where the output is structured data rather than documents.","whyChips":["Unlimited members","1 GB attachments","Self-hostable"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"grist","title":"Grist","provider":"Grist","planName":"freemium","logo":"/logos/grist.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Cloud Free plan: unlimited members, 2 guests per doc, 1 GB attachment storage per doc. Self-hosting is $0 for individuals and orgs with under $1M total annual income (free activation key).","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"4 level","display":"4 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"5 level","display":"5 level"},{"key":"workspace_automation_integration_readiness","label":"Auto","value":"4 level","display":"4 level"}],"url":"https://www.getgrist.com/"}},{"slug":"clickup","headline":"Best broad free team workspace","description":"ClickUp is the side pick because its Free Forever plan covers tasks, docs, whiteboards, and common project views in one place. It is broad enough to act as a lightweight operating system for a small project team, with collaborative docs, list and board views, and real-time collaboration all included without a credit card. The storage cap of 60 MB is the honest trade-off.","whyChips":["Tasks + docs","Unlimited members","No card"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"clickup","title":"ClickUp","provider":"ClickUp","planName":"freemium","logo":"/logos/clickup-business.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free Forever plan: unlimited tasks and unlimited free-plan members with collaborative Docs, whiteboards, and list/board/calendar views, capped at 60 MB of file storage, 5 Spaces, and 100 uses of advanced features.","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"4 level","display":"4 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"5 level","display":"5 level"},{"key":"workspace_automation_integration_readiness","label":"Auto","value":"3 level","display":"3 level"}],"url":"https://clickup.com"}},{"slug":"nuclino","headline":"Fast wiki-style team knowledge","description":"Nuclino is the side pick for teams that want a fast, no-clutter shared wiki with real-time collaboration and visual graph navigation, all free for unlimited users with up to 50 items and 2 GB total storage. The hard item cap of 50 means it fits small teams and focused projects rather than large knowledge bases, but for agent-assisted research and documentation it is a clean tool.","whyChips":["Unlimited users","2 GB storage","Real-time collab"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"nuclino","title":"Nuclino","provider":"Nuclino","planName":"freemium","logo":"/logos/nuclino-business.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free forever for unlimited users with up to 50 items (docs/pages) and 2 GB total storage, including real-time collaboration, all views, and 30-day version history.","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"4 level","display":"4 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"5 level","display":"5 level"}],"url":"https://www.nuclino.com"}}],"hero":{"slug":"grist","headline":"The relational database workspace","description":"Grist is the hero workspace pick because it is a spreadsheet-database hybrid that non-engineers can use fluently and engineers can extend with Python. The Cloud Free plan allows unlimited members and 2 guests per doc with 1 GB attachment storage per doc, and self-hosting is free for individuals and orgs with under $1M annual income. That combination makes it a durable, self-hostable foundation for agent-run projects where the output is structured data rather than documents.","whyChips":["Unlimited members","1 GB attachments","Self-hostable"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"grist","title":"Grist","provider":"Grist","planName":"freemium","logo":"/logos/grist.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Cloud Free plan: unlimited members, 2 guests per doc, 1 GB attachment storage per doc. Self-hosting is $0 for individuals and orgs with under $1M total annual income (free activation key).","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"4 level","display":"4 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"5 level","display":"5 level"},{"key":"workspace_automation_integration_readiness","label":"Auto","value":"4 level","display":"4 level"}],"url":"https://www.getgrist.com/"}},"nominations":[{"slug":"cryptpad","label":"Encrypted","description":"CryptPad is the nomination for teams that need end-to-end encryption as a first-class property — the server and operator cannot read stored documents. A free registered account gives 1 GB of encrypted storage across all apps (docs, sheets, slides, code, kanban, forms, whiteboard, calendar) with real-time collaboration. Per-file uploads are capped at 25 MB on the free plan.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"cryptpad","title":"CryptPad","provider":"CryptPad","planName":"freemium","logo":"/logos/cryptpad.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free registered account on cryptpad.fr gives 1 GB of end-to-end-encrypted storage with full access to every app (docs, sheets, slides, code/markdown, kanban, forms, whiteboard, calendar), real-time collaboration, teams, and shared folders, with per-file uploads capped at 25 MB.","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"3 level","display":"3 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"1 level","display":"1 level"}],"url":"https://cryptpad.fr/"}},{"slug":"affine","label":"Open-source","description":"AFFiNE is the nomination for teams that want a FOSS-leaning, local-first workspace with a Notion + Miro hybrid feature set. The free cloud plan includes unlimited local workspaces, 10 GB cloud storage, and up to 3 members per workspace, and the MIT-licensed FOSS edition is fully self-hostable — which means agent data never needs to leave the team's own infrastructure.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"affine","title":"AFFiNE","provider":"AFFiNE","planName":"freemium","logo":"/logos/affine.png","pricingModel":"freemium","accessBucket":"free","freeLimitText":"Free forever: unlimited local workspaces, 10 GB cloud storage, up to 3 members per workspace, 7-day cloud version history. FOSS edition fully self-hostable under MIT.","keyMetrics":[{"key":"workspace_content_api_access","label":"Content","value":"3 level","display":"3 level"},{"key":"workspace_collaboration_seats","label":"Seats","value":"2 level","display":"2 level"}],"url":"https://affine.pro/"}}],"paidPicks":[]},"domains":{"collection":"infrastructure","category":"domains","updatedAt":"2026-07-10","intro":"Free DNS, dynamic DNS, and provider subdomain options for agent-accessible hostnames.","picks":[{"slug":"duckdns","headline":"Simple dynamic DNS for free subdomains","description":"Duck DNS is the hero domain pick because it is the lowest-friction way to get a public hostname that tracks a changing home or office IP address. Token-based updates mean the update script can live in a router, a small device, or an agent cron without storing credentials in plaintext. The trade-off is that you only get *.duckdns.org subdomains — no custom apex domains — and login options have narrowed over time.","whyChips":["No card","Token updates","Free subdomain"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"duckdns","title":"Duck DNS","provider":"Duck DNS","planName":"free_forever","logo":"/logos/duckdns.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Free: multiple *.duckdns.org subdomains with token-based dynamic IP updates.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"5 domains","display":"5 domains"},{"key":"ddns_update_interval","label":"DDNS refresh","value":"5 min","display":"5 min"}],"url":"https://www.duckdns.org/"}},{"slug":"cloudns","headline":"Feature-rich free DNS","description":"ClouDNS is the side pick for developers who want more than basic A records from their free DNS provider. The Free Forever plan includes 1 DNS zone, 50 DNS records, 500,000 DNS queries per month on 4 unicast nameservers, plus 1 mail forward and 1 dynamic DNS hostname. That quota handles personal projects and small services without immediately hitting limits.","whyChips":["1 zone free","500k queries/mo","Dynamic DNS"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"cloudns","title":"ClouDNS","provider":"ClouDNS","planName":"freemium","logo":"/logos/cloudns-ddos-protected-s.png","pricingModel":"freemium","accessBucket":"conditional","freeLimitText":"Free Forever: 1 DNS zone, 50 DNS records, 500,000 DNS queries/month on 4 unicast nameservers, plus 1 mail forward and 1 dynamic DNS hostname.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"1 domains","display":"1 domains"}],"url":"https://www.cloudns.net/"}},{"slug":"dynv6","headline":"Unlimited free hosts with DNSSEC","description":"dynv6 is the side pick for self-hosters and home-lab users who want free public hostnames under dynv6.net, dns.army, or v6.rocks with full DNS record management and DNSSEC at no cost. The service is entirely free for private use with unlimited hosts, which makes it practical for agent fleets and IoT setups where each instance needs its own resolvable hostname.","whyChips":["Unlimited hosts","DNSSEC","No card"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"dynv6","title":"dynv6","provider":"dynv6","planName":"free_forever","logo":"/logos/dynv6.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Completely free for private use: unlimited hosts with freely chosen names under dynv6.net/dns.army/v6.rocks etc. (or delegate your own domain), with full DNS record management and DNSSEC at no cost.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"∞","display":"∞"}],"url":"https://dynv6.com/"}}],"hero":{"slug":"duckdns","headline":"Simple dynamic DNS for free subdomains","description":"Duck DNS is the hero domain pick because it is the lowest-friction way to get a public hostname that tracks a changing home or office IP address. Token-based updates mean the update script can live in a router, a small device, or an agent cron without storing credentials in plaintext. The trade-off is that you only get *.duckdns.org subdomains — no custom apex domains — and login options have narrowed over time.","whyChips":["No card","Token updates","Free subdomain"],"verified":{"verifiedAt":"2026-07-10","confidence":"high"},"offer":{"slug":"duckdns","title":"Duck DNS","provider":"Duck DNS","planName":"free_forever","logo":"/logos/duckdns.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Free: multiple *.duckdns.org subdomains with token-based dynamic IP updates.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"5 domains","display":"5 domains"},{"key":"ddns_update_interval","label":"DDNS refresh","value":"5 min","display":"5 min"}],"url":"https://www.duckdns.org/"}},"nominations":[{"slug":"sslip-io","label":"No signup","description":"sslip.io resolves any IP-embedded hostname instantly with no account and no credit card — for example, 192-168-0-1.sslip.io resolves to 192.168.0.1. It is the nomination for developers who need valid hostnames with built-in TLS certificates for local servers, Kubernetes ingress, and demos without registering a domain at all.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"sslip-io","title":"sslip.io","provider":"sslip.io","planName":"free_forever","logo":"/logos/sslip-io.png","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Unlimited resolution of any IPv4/IPv6-embedded hostname (e.g. 192-168-0-1.sslip.io) at no cost, with no account and no card; optional self-hosted branded subdomains via NS delegation.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"∞","display":"∞"}],"url":"https://sslip.io/"}},{"slug":"pages-dev","label":"Free subdomains","description":"Cloudflare pages.dev is the nomination for frontend-focused agent deployments where the agent's output is a static site or web app. It provides a free *.pages.dev subdomain with automatic TLS and global CDN, requiring nothing more than a GitHub repository connection.","verified":{"verifiedAt":"2026-07-10","confidence":"high"},"icon":"medal","offer":{"slug":"pages-dev","title":"Cloudflare pages.dev","provider":"Cloudflare pages.dev","planName":"free_forever","logo":"/logos/cloudflare.svg","pricingModel":"free_forever","accessBucket":"free","freeLimitText":"Free provider subdomain.","keyMetrics":[{"key":"free_domains","label":"Free domains","value":"∞","display":"∞"},{"key":"dns_records","label":"DNS records","value":"∞","display":"∞"}],"url":"https://developers.cloudflare.com/pages/"}}],"paidPicks":[]}},"partial":true}