Vercel Queues
Durable, serverless message queue for deferring async work: publish messages to topics and process them via managed consumer groups with automatic scaling, at-least-once delivery, and built-in retries.
- Card required
- No
Vercel Queues scores 5/10 for free value in Schedulers. The main constraint: 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.
Free value score
5.0 / 10Show the math
Key metrics
What you get
1M API operations/month free on Hobby
At-least-once delivery with automatic retries
Fan-out to multiple independent consumer groups
Push and poll consumption modes via @vercel/queue
Gotchas
- 1M API operations/month free
The Hobby plan includes the first 1,000,000 Queue API operations per month at no cost; the five billable operation types are Send, Receive, Delete, Visibility change, and Notify.
- Public beta
Vercel Queues is in public beta (introduced Feb 27, 2026) and uses the experimental queue/v2beta trigger type; the API, SDK, and pricing may change before general availability.
- Operations metered in 4 KiB chunks
Messages are billed in 4 KiB chunks, so a single 12 KiB message counts as three operations — large payloads consume the free quota faster.
- 2x billing on idempotency and max-concurrency
Sends that include an idempotency key and push deliveries that use max concurrency are billed at 2x units for that operation.
- Free quota does not roll over
The 1M monthly included operations reset each month; unused operations do not carry forward.
- Hobby is noncom only
The Hobby plan is intended for personal, noncom projects; commercial workloads require upgrading to a paid plan.
- 100 MB max message size
Individual messages cannot exceed 100 MB; combined with 4 KiB metering this means a max-size message is extremely operation-heavy.
- 7-day max retention
Message retention (TTL) ranges from 60 seconds to a maximum of 7 days, defaulting to 24 hours; messages expire and are dropped after their TTL.
- Push-mode compute billed separately
Functions invoked by Queues in push mode are charged at existing Fluid compute rates, in addition to the queue API operation cost.
Sources
-
"Managed Infrastructure hobby and pro resources | Resource | Hobby Included ... | Queue API Operations | First 1,000,000 | N/A | Regional"
https://vercel.com/docs/queues/pricing -
"Vercel Queues | Beta | API Operations | 1M / month included | Starting at $0.0000006 per operation | Custom"
https://vercel.com/pricing -
"Messages are metered in 4 KiB chunks. For example, a 12 KiB message counts as three operations. Sends with an idempotency key and push deliveries with max concurrency are billed at 2x units for that operation."
https://vercel.com/docs/queues/pricing -
"Vercel Queues is a durable event streaming system built with Fluid compute, and is now available in public beta for all teams."
https://vercel.com/changelog/vercel-queues-now-in-public-beta
