Synthetic Pack

LLM 6.6/10

GLM-5.2 (512k) + Kimi-K2.7-Code + MiniMax M3 + GLM-4.7-Flash + gpt-oss-120b flat-rate: 500 requests/5h + $24 API credits/week per pack, limits regenerate continuously (no lockouts); native Anthropic + OpenAI endpoints; stack packs to scale linearly

Requests / month: 72K
Concurrency: 1

Synthetic Pack scores 6.6/10 for free value in LLM. The main constraint: Paid — $30/mo per pack ($1/day); 2 packs = $60/mo with 2x limits + concurrency. GLM-5.2 (512k) + Kimi-K2.7-Code + MiniMax M3 + GLM-4.7-Flash + gpt-oss-120b flat-rate: 500 requests/5h + $24 API credits/week per pack, limits regenerate continuously (no lockouts); native Anthropic + OpenAI endpoints; stack packs to scale linearly.

Category LLM Kind Paid plan Caveats 1 concurrent request per model per pack — heavy same-model parallelism requires multiple packs; requests weighted by model price (GLM-5.2 = 1 req, GLM-4.7-Flash ~0.1 req); the $24/week credit is the real ceiling

Free value score

6.6 / 10

Show the math

Model Quality 9.2/10

value = 51 artificial_analysis_score

weighted score = 0.2 * 9.2/10

Requests 4.7/10

value = 72000 req/mo

weighted score = 0.1 * 4.7/10

Concurrency 1/10

value = 1 parallel

score = 1/10

Access 10/10

value = 10/10

score = 10/10

Price Efficiency 6.1/10

value = 3.47 value/$

weighted score = 0.7 * 6.1/10

Key metrics

GLM-5.2 51 Best model

72K R/mo Requests/month

1 concurrent Concurrency

3.4667x Value ratio

What you get

Paid — $30/mo per pack ($1/day); 2 packs = $60/mo with 2x limits + concurrency. GLM-5.2 (512k) + Kimi-K2.7-Code + MiniMax M3 + GLM-4.7-Flash + gpt-oss-120b flat-rate: 500 requests/5h + $24 API credits/week per pack, limits regenerate continuously (no lockouts); native Anthropic + OpenAI endpoints; stack packs to scale linearly

~$104 of usage per month in API credits + 500 req/5h for $30 => ~3.5x face value on GLM-5.2/K2.7 with real dual API for custom agents

Gotchas

No gotchas documented yet. Contribute one if you know a constraint we missed.

Sources

"Each pack: 500 requests per 5 hours, $24 of API credits per week, 1 concurrent request per model; limits regenerate continuously"
https://synthetic.new/pricing
"zai-org/GLM-5.2 (512k context) and moonshotai/Kimi-K2.7-Code included flat-rate"
https://synthetic.new/pricing

paid value llm-providers

Open service ← All LLM

Alternatives

Same category · similar quota

Synthetic High-Volume (2 Packs)Synthetic High-Volume (2 Packs) · paid · Paid plan

6.6

Two stacked packs: 1,000 requests/5h + $48 API credits/week + 2 concurrent requests per model across GLM-5.2 (512k), Kimi-K2.7-Code, MiniMax M3 and other open models; native Anthropic + OpenAI endpoints

3.5×51.1 AA3.5×

May be outdatedChecked 23d ago

Flat ratePaid only

NVIDIA NIMNVIDIA NIM · freemium · Hosted model API

6.5

Free OpenAI-compatible API access to a large catalog of hosted open models on NVIDIA's developer platform.

44.4 AA40/min

May be outdatedChecked 28d ago

NoncomEval

OVHcloud AI EndpointsOVHcloud AI Endpoints · freemium · Hosted LLM inference API (EU, free anonymous tier)

6.2

OVHcloud AI Endpoints offers a permanent free anonymous tier — no signup or API key — to call 40+ EU-hosted open-weight models (Llama, Qwen, Mixtral, DeepSeek-R1 distill) via an OpenAI-compatible API.

37 AA2/min

Checked 10d ago

No signup