north-mini-code-1.0

north-mini-code-1.0

8,299 Downloads Updated 6 days ago

North Mini Code is Cohere's first model for developers — a 30B Mixture-of-Experts model with 3B active parameters, built for agentic software engineering.

Cancel

tools thinking

CLI cURL Python JavaScript

Documentation Documentation

ollama run north-mini-code-1.0

curl http://localhost:11434/api/chat \ -d '{ "model": "north-mini-code-1.0", "messages": [{"role": "user", "content": "Hello!"}] }'

from ollama import chat

response = chat( model='north-mini-code-1.0', messages=[{'role': 'user', 'content': 'Hello!'}], ) print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({ model: 'north-mini-code-1.0', messages: [{role: 'user', content: 'Hello!'}], }) console.log(response.message.content)

Applications

Claude Code `ollama launch claude --model north-mini-code-1.0`

Codex App `ollama launch codex-app --model north-mini-code-1.0`

OpenClaw `ollama launch openclaw --model north-mini-code-1.0`

Hermes Agent `ollama launch hermes --model north-mini-code-1.0`

Codex `ollama launch codex --model north-mini-code-1.0`

OpenCode `ollama launch opencode --model north-mini-code-1.0`

Models

View all →

Name

7 models

Size / Usage

Context

Input

north-mini-code-1.0:latest 19GB · 488K context window · Text · 6 days ago

north-mini-code-1.0:latest

19GB

488K

Text

north-mini-code-1.0:q4\_K\_M 19GB · 488K context window · Text · 6 days ago

north-mini-code-1.0:q4\_K\_M

19GB

488K

Text

north-mini-code-1.0:q8\_0 32GB · 488K context window · Text · 6 days ago

north-mini-code-1.0:q8\_0

32GB

488K

Text

north-mini-code-1.0:bf16 61GB · 488K context window · Text · 6 days ago

north-mini-code-1.0:bf16

61GB

488K

Text

north-mini-code-1.0:mlx-mxfp8 MLX 31GB · 488K context window · Text · 6 days ago

north-mini-code-1.0:mlx-mxfp8 MLX

31GB

488K

Text

north-mini-code-1.0:mlx-nvfp4 MLX 20GB · 488K context window · Text · 6 days ago

north-mini-code-1.0:mlx-nvfp4 MLX

20GB

488K

Text

north-mini-code-1.0:mlx-bf16 MLX 61GB · 488K context window · Text · 6 days ago

north-mini-code-1.0:mlx-bf16 MLX

61GB

488K

Text

Readme

North Mini Code is the first model in Cohere’s new family of models, and is specifically designed and trained for agentic software engineering tasks.

• Agentic coding focus, post-trained with two-stage supervised fine-tuning followed by reinforcement learning with verifiable rewards (RLVR) on real-world software engineering and terminal tasks.
• 256K context length with up to 64K output tokens, optimized for repository-scale understanding and long-horizon agent trajectories.
• Trained across multiple agent harnesses (SWE-Agent, mini-SWE-agent, OpenCode, Terminus 2) for robustness in real-world tooling environments rather than a single scaffold.
• Native tool-use and interleaved thinking support, designed to plug into coding agents like OpenCode.

On Artificial Analysis’ Coding Index, North Mini Code scores 33.4, outperforming similarly sized open models like Qwen3.5 (35B-A3B), Gemma 4 (26B-A4B), and Devstral Small 2 (24B), as well as substantially larger models including Nemotron 3 Super (120B-A12B), Mistral Small 4 (119B-A6B), and Devstral 2 (123B).

Architecture

North Mini Code is a decoder-only Transformer-based sparse Mixture-of-Experts model. It interleaves sliding-window attention (with RoPE) and global attention (with no positional embeddings) in a 3:1 ratio. The feed-forward block is an MoE block with 128 experts, 8 of which are activated per token, each using SwiGLU activation. The router applies a sigmoid activation before top-k selection, and a single dense layer precedes the sparse layers.

Tool use

North Mini Code is trained for tool use and agentic coding, and supports interleaved thinking — it works best with thinking enabled. For best performance, pass model-generated thinking content forward to subsequent agentic steps and chat turns. Tool descriptions are best provided as JSON schema.

License

North Mini Code is released under the Apache 2.0 license, and also requires adhering to Cohere Lab’s Acceptable Use Policy.

Reference

• Blog post
• Hugging Face

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)