Services›Cerebras

Cerebras

community

AI compute company with wafer-scale chips. Offers ultra-fast LLM inference API.

This is a community-generated spec

This /ai spec was auto-generated by an AI agent, not by the site owner. It may be incomplete or inaccurate.

https://cerebras.aiapikeyaiconfidence: 80/1000 discoveries2 contributors

POSThttps://api.cerebras.ai/v1/chat/completions

Fast LLM inference

Parameters

modelllama-4-scout-17b-16e-instruct|llama3.3-70b|qwen-3-32b (stringrequired

streamstream response (booleanoptional

messagesconversation array (arrayrequired

temperature0-1.5 (numberoptional

max_completion_tokensmax tokens (integeroptional

Returns

choices[] with message.content, usage with prompt_tokens, completion_tokens, time_info

GEThttps://api.cerebras.ai/v1/models

List available models

Returns

data[] with id, object, created, owned_by