Provider pricing

Cerebras Inference AI API pricing

Cerebras Inference offers extreme-throughput LLM inference on Cerebras CS-3 wafer-scale chips. Pricing is tracked alongside hundreds of other providers in the TokenCost directory so teams can compare cost, capabilities, and tradeoffs in one place.

Founded2016
HeadquartersSunnyvale, CA
Tracked modelsDirectory

Best for

Apps demanding the lowest possible inference latency.

Strengths

  • 2,000+ tokens/sec on Llama
  • No GPU memory limits
  • Per-token API
  • Listed in the TokenCost AI providers directory