Groq LPU AccelerationGroq LPU 极速推理
World-record LPU inference (500-800 t/s). Generous permanent free tier with 30 RPM.
Core Free Tier
永久免费层 30 RPM / 14,400 RPD永久免费
Permanent free tier with 30 RPM and 14,400 daily requests.
Reset Cycle: 每日刷新
Eligibility & Verification Requirements
Phone VerificationNot verified
Credit CardNot verified
Network / RegionNot verified
Concurrency RateRPM Not verified · TPM Not verified
Operation status and evidence
chat_completionsProvider-claimed; documentation not checked- Protocol
- openai
- Checked
- Not checked
- Endpoint
https://api.groq.com/openai/v1/chat/completions- Evidence
- Not provided
llama-3.3-70b-versatilellama-3.1-8b-instantdeepseek-r1-distill-llama-70bIntegration
No OpenAI Chat Completions example can currently be generated safely. For generic or non-chat operations, follow the official documentation ↗ rather than adapting a chat endpoint.
Gotchas & Risk Precautions
- Strict TPM limit applies on free tier.
