Skip to content
ShilpBot
AI Models

Cohere North Mini Code (Free): Specs, Limits and How to Use It

Cohere North Mini Code free: sparse mixture-of-experts agentic coding model, text input, 256K context. Specs, limits, test prompts and how to try it.

By ShilpBot Team · Published · 3 min read

North Mini Code is described as Cohere's first agentic coding model and the debut of its North family. It is a sparse mixture-of-experts model with 30 billion total and 3 billion active parameters.

It supports a wide range of sampling controls, including top-k, frequency penalty and presence penalty, which gives you more control over output style than some other free models.

Specifications checked against the public OpenRouter catalogue on 2026-11-16. See the live model page for current limits.

Cohere North Mini Code quick facts

DetailValue
DeveloperCohere
Model IDcohere/north-mini-code:free
Added to the catalogue2026-06-17
TypeSparse mixture-of-experts agentic coding model
Parameters3B active / 30B total
Context window256K tokens (256,000)
Maximum output64,000 tokens
Input typestext
PriceFree

Feature support from the catalogue: Tool calling: listed; JSON / structured output: not listed; Reasoning: supported, effort not adjustable.

What Cohere North Mini Code is good for

  • Coding-agent workflows where you want fine control over sampling settings.
  • Mid-sized refactors that fit in a 256K context window.
  • Teams already using Cohere who want to compare it with other vendors.

Limits of Cohere North Mini Code

  • Text only, with no structured-output mode listed.
  • The catalogue shows only a free listing for it.
  • First-generation models in a new family can have rough edges. Test before relying on it.

How Cohere North Mini Code compares

  • Recursive Flint: It has a smaller context window (256,000 against 262,144 tokens); its output cap is 64,000 tokens against 235,929; it activates 3B parameters per token against 2.5B.
  • Particle AI Particle-1: It has a smaller context window (256,000 against 1,048,576 tokens); its output cap is 64,000 tokens against 131,072; it does not accept image input.
  • NVIDIA Nemotron 3.5 Lightning: It has a smaller context window (256,000 against 1,000,000 tokens); its output cap is 64,000 tokens against 65,536; it activates 3B parameters per token against 3B.

Calling Cohere North Mini Code from code

Use the model ID cohere/north-mini-code:free with any OpenAI-compatible client pointed at OpenRouter:

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "cohere/north-mini-code:free", "messages": [{"role": "user", "content": "Explain what you can do in two sentences."}]}'

Prompts that show what Cohere North Mini Code can do

  1. Ask it to convert a callback-based function to async/await and explain what changed.
  2. Give it a failing test and the code under test, and ask for a fix.
  3. Vary the top-k setting on the same prompt and compare the diversity of answers.

What Cohere North Mini Code costs

The catalogue lists only a free version of this model, with no paid listing to fall back on. If the free tier changes, you will need to switch to a different model. Free pricing is set by the provider and can change at any time, so check the model page before you plan around it.

Keep reading

All free models are compared in our free AI chat models roundup. New to terms like mixture-of-experts? Start with how neural networks learn.

Frequently asked questions

What is the North family from Cohere?

The listing describes North Mini Code as the debut of Cohere's North family and its first agentic coding model.

Is Cohere North Mini Code free, and what are its limits?

As of 2026-11-16 the catalogue lists it at zero price. 256K tokens (256,000) context window, up to 64,000 output tokens, model ID cohere/north-mini-code:free. Free use can be rate limited and providers may change it.

#Cohere North Mini Code#Cohere free model#free AI model#AI chat model#AI coding model#Kilo Gateway#OpenRouter