# MiniMax M3 — Open-Weight Frontier Coding, 1M-Token Context, and Native Multimodality on a New Sparse-Attention Architecture

> M3 packages frontier coding, a million-token context, and native multimodality into one open-weight model.

- Org: MiniMax
- Date: 2026-06-01
- Category: model
- Source: https://www.minimax.io/blog/minimax-m3
- Canonical page: https://ai-tldr.dev/releases/minimax-m3/

## In short

MiniMax shipped M3, an open-weight successor to M2.7 built on a new MiniMax Sparse Attention design. It scores 59.0% on SWE-Bench Pro, 66.0% on Terminal-Bench 2.1, and serves a 1M-token context window.

## What it is

M3 is the next model in the MiniMax line after M2.7. It targets coding and agentic work but is natively multimodal — image, video, and desktop operation are trained in from step zero. MiniMax calls it the first open-weight model that combines frontier coding scores, a 1M-token context window, and multimodal input in a single checkpoint.

## How it works

M3 uses MiniMax Sparse Attention (MSA), a new attention design that pairs softmax expressiveness with a top-k block selection step, keeping sub-quadratic complexity at long contexts. MiniMax reports >9x prefill speedup and >15x decoding speedup at 1M-token context versus M2, with roughly 1/20th the per-token compute. The API, MiniMax Code, and Token Plans go live today; the technical report and open weights are due on Hugging Face and GitHub within about ten days.

## Why it matters

Open-weight frontier coding models with real 1M-token context have been thin on the ground, and most run text-only. If the weights ship as advertised, M3 gives self-hosters a single model that handles long-repo coding, computer use, and multimodal grounding without stitching together specialised components.

## Who it's for

Open-weight users, agentic-coding builders, long-context power users

## Try it

platform.minimax.io — Plus plan: ~1.7B tokens / $20/mo

## Links

- [MiniMax Blog](https://www.minimax.io/blog/minimax-m3)
- [Model Page](https://www.minimax.io/models/text/m3)
- [GitHub Repo](https://github.com/MiniMax-AI/MiniMax-M3)
- [MSA Paper (arXiv)](https://arxiv.org/abs/2606.13392)
- [MSA Code](https://github.com/MiniMax-AI/MSA)
- [MSA Explainer](https://huggingface.co/blog/AtlasCloud-AI/minimax-goes-sparse)
- [API Coding Docs](https://platform.minimax.io/docs/guides/text-ai-coding-tools)