all tiers / Groq
GroqCloud API free tier limits (2026)
free plan with per-model limits. gpt-oss and qwen models get 30 rpm / 1k requests a day.
per minute30 RPM, 8K TPM (gpt-oss-120b, gpt-oss-20b, qwen3.8-27b)
per day1K RPD, 200K TPD (same models)
per month
other limits / notes from sourcegpt-oss-120b / gpt-oss-20b / qwen3.8-27b: 30 RPM, 1K RPD, 8K TPM, 200K TPD; gpt-oss-safeguard-20b: 3 RPM, 1K RPD, 2K TPM, 200K TPD; prompt-guard-2: 30 RPM, 14.4K RPD, 15K TPM, 500K TPD. audio: whisper-large-v3(-turbo): 20 RPM, 2K RPD, 7.2K audio-s/h, 28.8K audio-s/day; orpheus (v1-english, arabic-saudi): 10 RPM, 100 RPD, 1.2K TPM, 3.6K TPD
card required
phone required
used for trainingno (services agreement 4.2)
age req (as stated by the source)18+
requirementsnot for consumer use
region restrictions
not verified: docs label the table 'base limits for the Developer plan'; free-plan applicability unclear
corrected 2026-10-18: added safeguard and prompt-guard-2 limits. no training on your data (services agreement 4.2). the docs call the table 'base limits for the developer plan'. source
models included
openai/gpt-oss-120bopenai/gpt-oss-20bopenai/gpt-oss-safeguard-20bqwen/qwen3.8-27bwhisper-large-v3whisper-large-v3-turbocanopylabs/orpheus-v1-englishcanopylabs/orpheus-arabic-saudimeta-llama/llama-prompt-guard-2-22mmeta-llama/llama-prompt-guard-2-86m
notes
limits are org-level; cached tokens don't count. values from the 'Free Plan Limits' tab.
couldn't find these on the official page: card required, phone required, region restrictions. if you can, send the link and we'll check it.
sources
- https://console.groq.com/docs/rate-limits
- https://console.groq.com/docs/your-data
- https://console.groq.com/docs/legal/services-agreement