Cohere releases North Mini Code, an open-source coding model for developers
Original title:Introducing North Mini Code: Cohere’s First Model For Developers
AI summary
Cohere released North Mini Code, a 30B-parameter MoE model (3B active parameters) open-sourced under Apache 2.0. It scores 33.4 on the Artificial Analysis Coding Index, surpassing comparable models such as Qwen3.5 and Gemma 4. Post-training uses two-stage SFT and RLVR, reaching 80.2% pass@10 on SWE-Bench Verified and 55.1% on Terminal-Bench v2. It supports 64K/128K context lengths and is optimized for agentic coding tasks.
Key Takeaways
- Cohere released North Mini Code,a 30B-parameter MoE model (3B active parameters) open-sourced…
- It scores 33.
- 4 on the Artificial Analysis Coding Index,surpassing comparable models such as Qwen3.
- 5 and Gemma 4.
- Post-training uses two-stage SFT and RLVR,reaching 80.
Quick Facts
- Published
- Updated
- Source
- Hugging Face 社区博客(混合发现)
- Category
- Model Releases
- AI score
- 73
Cohere的新编码模型North Mini Code以30B参数MoE架构,在SWE-bench pass@1达到61%,Apache 2.0开源,是小模型在agent coding领域真正可用的信号。
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: Cohere releases North Mini Code, an open-source coding model for developers. June 9, 2026. https://new.nat6.net/en/news/cmq74dg8m006psloprllkcbmm/