Cerebras

Current free tier

What's free

Fastest inference around (~2,000 tokens/sec) on a rotating lineup of open-weight models (Llama, Qwen, GPT-OSS).

Limits

~1M tokens/day

The catch

Small model lineup that rotates often: a model you rely on can disappear from the free tier.

Sign-up

No card needed

See it in the full ledger · JSON · raw change data · RSS of all changes

Change history

• 2026-07-04

  What's free changed.
  Before: Fastest inference around (~2,000 tokens/sec) on Llama-family models.
  After: Fastest inference around (~2,000 tokens/sec) on a rotating lineup of open-weight models (Llama, Qwen, GPT-OSS).

• 2026-07-04

  The catch changed.
  Before: Small model selection compared to other providers.
  After: Small model lineup that rotates often: a model you rely on can disappear from the free tier.