Cohere · all models ›

CNorth Mini Code (free)OPEN

North Mini Code is Cohere's first agentic coding model and the debut of its North family.

256KContext window · tokens
FreeInput price · per M tokens
FreeOutput price · per M tokens
CohereProvider

Prices update automatically — checked daily against provider list prices.

See model comparisons → Compare all model prices

Benchmarks & performance

Independent benchmark scores for North Mini Code (free), measured by Artificial Analysis. Higher is better.

Intelligence index20.2
Coding index36.5
GPQA75.7%
Humanity's Last Exam11.1%
Long Context Reasoning36%
SciCode38.2%
IFBench57.6%
τ²-Bench37.4%
τ-Bench Banking6.4%
Terminal-Bench35.6%
Terminal-Bench Hard31.1%
⚡ Speed80.9 tokens/sec
⏱ Latency0.3s to first token
vs. models measured heretop 63%
Scores higher than 37% of the 273 models measured by Artificial Analysis and tracked here.
Benchmark data by Artificial Analysis

About this model

North Mini Code (free) is an open-weight AI model by Cohere. You can download and self-host it for free; the prices below are hosted-API list prices, tracked daily, for when you prefer convenience over self-hosting.

Frequently asked questions

What is North Mini Code (free)?

North Mini Code (free) is an AI language model from Cohere. It is open-weight: you can download it and run it on your own hardware, for free. It scores 20.2 on the Artificial Analysis intelligence index.

Is North Mini Code (free) free?

The weights are free and open — you can self-host North Mini Code (free) and pay nothing per token. If you prefer a hosted API, list prices are Free per million input tokens and Free per million output tokens.

What is North Mini Code (free) good at?

Independent benchmarks from Artificial Analysis give it GPQA 75.7%, Humanity's Last Exam 11.1%, Long Context Reasoning 36%, SciCode 38.2%, IFBench 57.6%, τ²-Bench 37.4%, τ-Bench Banking 6.4%, Terminal-Bench 35.6%, Terminal-Bench Hard 31.1%. It is particularly used for code generation.

How fast is North Mini Code (free)?

It generates about 80.9 tokens per second, with a median 0.3s delay before the first token. Measured independently by Artificial Analysis.

Can I self-host North Mini Code (free)?

Yes. North Mini Code (free) has open weights, so you can download it and run it on your own GPU or server with tools like Ollama, vLLM or llama.cpp — with no per-token cost.

Related models

Command R (08-2024)CohereCommand R7B (12-2024)Cohere