Cerebras Inference

permanent free · inference_host · verification: partially_verified (65%, doc_fetch, 2026-07-17) · setup guide →

The record

Base URLs	https://api.cerebras.ai/v1 openai-compatible
API key required	yes
Credit card	no
Phone verification	unknown
May train on prompts	unknown
Quota	free_tier [rpm, tpm, tpd] “Official rate-limits page (Free Trial tier, per model, fetched 2026-07-17): gpt-oss-120b / zai-glm-4.7 / gemma-4-31b: 5 RPM, 30K TPM, 1M TPH, 1M TPD. cheahjs README (older): gpt-oss-120b: 30 requests/minute, 60,000 tokens/minute, 900 requests/hour, 1,000,000 tokens/hour, 14,400 requests/day, 1,000,000 tokens/day.”
Signup friction	3 steps · ~5m to first key
Env var	CEREBRAS_API_KEY
Notes	Ultra-fast inference (~2,600 tok/s per seed repo). Ongoing free tier per both seed repos, though the official rate-limits page labels the tier 'Free Trial' — watch for category change to trial_credit if Cerebras starts expiring access.

model	capabilities	context	free	limits
gemma-4-31b	chat, vision	?	yes	5 rpm · 30000 tpm · 1000000 tpd
gpt-oss-120b	chat, reasoning	131,072	yes	5 rpm · 30000 tpm · 1000000 tpd
zai-glm-4.7	chat	131,072	yes	5 rpm · 30000 tpm · 1000000 tpd