Skip to main content
kRouter
All providers

Pay-per-token

cohere/

Cohere

Command R+ and embeddings.

cohere.com

Quick setup

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

  2. Go to Providers → click Cohere

  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
cohere/command-r-plus

How Cohere billing works with kRouter

Cohere 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. 3 models are routable as cohere/<model>.

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

Cohere models & pricing

These are the model ids kRouter routes to. This provider does not publish per-token rates.

ModelRoute as
Command R+ (Aug 2024)cohere/command-r-plus-08-2024
Command R (Aug 2024)cohere/command-r-08-2024
Command A (Mar 2025)cohere/command-a-03-2025

Cohere FAQ

Is Cohere free through kRouter?

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

How do I use Cohere with Claude Code or Cursor?

Connect Cohere 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 cohere/<model>.

Which Cohere models can kRouter route to?

3 models, including Command R+ (Aug 2024), Command R (Aug 2024), Command A (Mar 2025). Every one is addressed as cohere/<model> from any client.

What happens when Cohere 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.