Skip to main content
kRouter
All providers

Pay-per-token

cerebras/

Cerebras

Wafer-scale chip inference.

cerebras.ai

Quick setup

  1. Open the kRouter dashboard at http://localhost:20128

  2. Go to Providers → click Cerebras

  3. Paste your API key, click Save

  4. Test the connection

Use in your IDE

Point any OpenAI-compatible client at kRouter using your local API key.

ide configbash
# Endpoint
http://localhost:20128/v1

# API key (from Dashboard → API Keys)
sk-krouter-XXXX

# Model
cerebras/llama-4-405b

How Cerebras billing works with kRouter

Cerebras is pay-per-token. You bring your own API key and kRouter routes to it, so you keep your own rate limits and billing relationship. kRouter authenticates it via apikey. 6 models are routable as cerebras/<model>.

Through kRouter, Cerebras serves chat and code completion. Requests reach it through the same local OpenAI-compatible endpoint as every other provider, so switching to or from Cerebras is a one-line model change in your client rather than a rewrite.

Cerebras models & pricing

Prices are USD per 1M tokens, cheapest first. kRouter bills nothing on top; this is the provider's own rate.

ModelRoute asInput /1MOutput /1M
Qwen3 235B A22Bcerebras/qwen-3-235b-a22b-instruct-2507$0.50$2.00
Qwen3 32Bcerebras/qwen-3-32b$0.50$2.00
GPT OSS 120Bcerebras/gpt-oss-120b
ZAI GLM 4.7cerebras/zai-glm-4.7
Llama 3.3 70Bcerebras/llama-3.3-70b
Llama 4 Scoutcerebras/llama-4-scout-17b-16e-instruct

Cerebras links

Cerebras FAQ

Is Cerebras free through kRouter?

Cerebras is pay-per-token: you bring your own API key and are billed by Cerebras at their rates. kRouter adds nothing on top and is free and MIT-licensed.

How do I use Cerebras with Claude Code or Cursor?

Connect Cerebras in the kRouter dashboard, then point your tool at kRouter's local endpoint (http://localhost:20128/v1) with a kRouter API key. Any OpenAI-compatible client works, and models are addressed as cerebras/<model>.

Which Cerebras models can kRouter route to?

6 models, including GPT OSS 120B, ZAI GLM 4.7, Llama 3.3 70B. Every one is addressed as cerebras/<model> from any client.

What happens when Cerebras is rate limited?

kRouter detects the limit, parks that account until its real reset time, and fails over to your next connected account or provider — so the request still completes instead of erroring.