Groq

permanent free · inference_host · verification: partially_verified (70%, doc_fetch, 2026-07-17) · setup guide →

The record

Base URLs	https://api.groq.com/openai/v1 openai-compatible
API key required	yes
Credit card	no
Phone verification	no
May train on prompts	unknown
Quota	free_tier [rpm, rpd, tpm, tpd] “llama-3.3-70b-versatile: 30 RPM, 1K RPD, 12K TPM, 100K TPD; llama-3.1-8b-instant: 30 RPM, 14.4K RPD, 6K TPM, 500K TPD; openai/gpt-oss-120b: 30 RPM, 1K RPD, 8K TPM, 200K TPD (varies by model). "Rate limits apply at the organization level, not individual users."”
Signup friction	3 steps · ~5m to first key
Env var	GROQ_API_KEY
Notes	Ultra-fast LPU inference. Llama 4 Maverick deprecated. Model list retrievable at https://api.groq.com/openai/v1/models.

model	capabilities	context	free	limits
llama-3.1-8b-instant	chat, tool_calling	131,072	yes	30 rpm · 14400 rpd · 6000 tpm · 500000 tpd
llama-3.3-70b-versatile	chat, tool_calling	131,072	yes	30 rpm · 1000 rpd · 12000 tpm · 100000 tpd
meta-llama/llama-4-scout-17b-16e-instruct	chat, vision	131,072	yes	30 rpm · 1000 rpd · 30000 tpm · 500000 tpd
openai/gpt-oss-120b	chat, reasoning, tool_calling	131,072	yes	30 rpm · 1000 rpd · 8000 tpm · 200000 tpd
qwen/qwen3-32b	chat, reasoning	131,072	yes	60 rpm · 1000 rpd · 6000 tpm · 500000 tpd