Cohere Releases North Mini Code, a Small Open-Weight Model for Coding
Artificial Analysis·
Cohere released North Mini Code, a small 30B parameter (3B active) open weights coding model. This model achieves competitive coding performance for its size and speed, positioning it as a focused option in the open-weight ecosystem.
Cohere released North Mini Code, a new open weights model designed for coding, featuring 30 billion total parameters with 3 billion active parameters. This text-only model scored 27.6 on the Artificial Analysis Intelligence Index and 33.4 on the Artificial Analysis Coding Index. It operates with a 256k token context window and is available under the Apache 2.0 license.
Artificial Analysis Intelligence Index
27.6
Artificial Analysis Coding Index
33.4
Output Tokens per Second
207.9
Time to First Token
0.25s
Context Window
256k tokens
License
Apache 2.0
North Mini Code scored above gpt-oss-20B (high) on the Artificial Analysis Intelligence Index and outperformed GLM-4.7-Flash on the Artificial Analysis Coding Index. It also shows notable speed, achieving approximately 208 output tokens per second with a low time to first token of 0.25 seconds. However, the model underperforms on general agentic tasks, scoring lower on GDPval-AA and τ²-Bench Telecom, indicating its specialized focus.
North Mini Code is an open weights model, allowing for self-hosting or access via Cohere's API. Its Apache 2.0 license permits commercial use, making it a flexible option for developers. The model's small active parameter count and high speed make it suitable for applications requiring efficient, specialized coding capabilities.
Artificial Analysis Intelligence and Coding Index rankings comparing performance metrics across various small open weights models.
Cohere just released North Mini Code, a small 30B parameter (3B active) open weights coding model that scores 27.6 on the Artificial Analysis Intelligence Index
Less than a month since @cohere's last model release, Command A+, has launched another open weights model that is optimized for coding, and much smaller at 30B total parameters and 3B active parameters.
Key Takeaways:
➤ Achieves 27.6 on the Artificial Analysis Intelligence Index, above gpt-oss-20B (high) at 24.5 and just below Mistral Small 4 (119B parameters, 6.5B active) at 27.8
➤ Scores competitively on the Artificial Analysis Coding Index (weighted average of Terminal-Bench Hard and SciCode) against open weights models in its size class, scoring 33.4, significantly above GLM-4.7-Flash at 25.9, and below Qwen3.6 35B A3B at 35.2. However, it underperforms on non-coding agentic tasks, scoring 14% on GDPval-AA and 37% on 𝜏²-Bench Telecom
➤ On Cohere’s API, North Mini Code is faster than several comparable open weights models of its intelligence and size class (~199 output tokens per second)
➤ North Mini Code is a text-only 30B total parameter and 3B active parameter model, and is open-sourced under the Apache 2.0 license
