MiniMax M3: frontier coding, 1M-token context and native multimodal in one model
Original title:MiniMax M3: Frontier Coding, 1M Context, Native Multimodality — All in One Model
AI summary
MiniMax M3 is an open-source frontier model with advanced coding and AI agent capabilities. It supports a 1M-token ultra-long context window and adopts a new sparse attention architecture called MSA (MiniMax Sparse Attention). This architecture cuts per-token compute cost at 1M-token context to 1/20 of the previous generation, improves prefill speed by more than 9x and decode speed by more than 15x. On the SWE-Bench Pro coding benchmark, MiniMax M3 scores 59.0%, surpassing GPT-5.5 and Gemini 3.1 Pro and approaching Opus 4.7. The model is available through MiniMax Code, Token Plan and API services.
Key Takeaways
- MiniMax M3 is an open-source frontier model with advanced coding and AI agent capabilities.
- It supports a 1M-token ultra-long context window and adopts a new sparse attention architecture…
- This architecture cuts per-token compute cost at 1M-token context to 1/20 of the previous…
- On the SWE-Bench Pro coding benchmark,MiniMax M3 scores 59.
- 0%,surpassing GPT-5.
Quick Facts
- Published
- Updated
- Source
- MiniMax:Blog(网页)
- Category
- Model Releases
- AI score
- 83
MiniMax M3 把开源模型的编码能力推到了 GPT-5.5 和 Opus 4.7 这条线上,还附带 1M 上下文和原生多模态,这是开源社区真正能打的前沿选项,做 Agent 的值得立刻跑一下。
Related entities
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: MiniMax M3: frontier coding, 1M-token context and native multimodal in one model. June 1, 2026. https://new.nat6.net/en/news/cmpunu0g50002slbrlfonpoc3/