MiniMaxAI/MiniMax-M3
<div align="center"> <img width="60%" src="figures/logo.svg" alt="MiniMax"> </div> <hr>
<p align="center"> <a href="https://agent.minimax.io/" target="blank"><img src="https://img.shields.io/badge/MiniMax%20Agent-FF6C37?logo=minimax&logoColor=white" alt="MiniMax Agent"></a> <a href="https://platform.minimax.io/docs/guides/text-generation" target="blank"><img src="https://img.shields.io/badge/API-FF6C37?logo=minimax&logoColor=white" alt="API"></a> <a href="https://www.minimax.io" target="blank"><img src="https://img.shields.io/badge/MiniMax%20Website-FF6C37?logo=minimax&logoColor=white" alt="MiniMax Website"></a> <br> <a href="https://modelscope.cn/organization/minimax" target="blank" rel="noopener noreferrer"><img alt="ModelScope MiniMax AI" src="https://img.shields.io/badge/ModelScope-MiniMax%20AI-white?labelColor=%23EF3D5D"></a> <a href="https://platform.minimaxi.com/docs/faq/contact-us" target="blank"><img src="https://img.shields.io/badge/WeChat-07C160?logo=wechat&logoColor=white" alt="WeChat"></a> <a href="https://discord.com/invite/DPC4AHFCBw" target="blank"><img src="https://img.shields.io/badge/Discord-5865F2?logo=discord&logoColor=white" alt="Discord"></a> <a href="https://huggingface.co/MiniMaxAI" target="blank"><img src="https://img.shields.io/badge/Hugging%20Face-FFD21E?logo=huggingface&logoColor=black" alt="Hugging Face"></a> <a href="https://github.com/MiniMax-AI/MiniMax-M3" target="blank"><img src="https://img.shields.io/badge/GitHub-181717?logo=github&logoColor=white" alt="GitHub"></a> <a href="https://arxiv.org/abs/2606.13392" target="blank"><img src="https://img.shields.io/badge/arXiv-2606.13392-B31B1B?logo=arxiv&logoColor=white" alt="arXiv Paper"></a> <a href="https://huggingface.co/MiniMaxAI/MiniMax-M3/blob/main/LICENSE" target="blank"><img src="https://img.shields.io/badge/LICENSE-4CAF50?logo=creativecommons&logoColor=white" alt="LICENSE"></a> </p>
MiniMax-M3 is a native multimodal model with 1M context. It has ~428B parameters and ~23B activated parameters.
Highlights:
- Native Multimodality: M3 undergoes mixed-modality training from the very first step, enabling deeper semantic fusion across text, image, and video.
- Context Scaling via Sparse Attention: M3 introduces MiniMax Sparse Attention (MSA) to improve long context efficiency. M3 delivers 9× prefill and 15× decode speedups compared to M2 at 1M context, reducing per-token compute to 1/20.
- Coding & Cowork Capability: M3 achieves frontier-level performance across long-horizon agentic benchmarks, excelling in both coding and cowork.
<p align="center"> <img width="100%" src="figures/benchmark.jpeg"> </p>
MiniMax Sparse Attention (MSA)
M3 is powered by **MiniMax Sparse Attention (MSA)**, a high-performance sparse attention operator designed for million-token contexts. Compared with GQA, MSA dramatically reduces the attention compute and memory footprint while preserving model quality.
<p align="center"> <img width="100%" src="figures/efficiencygqavs_msa.png" alt="GQA vs MSA Efficiency Comparison"> </p>
📄 Read the technical report: arXiv:2606.13392 · Hugging Face Papers
How to Use
M3 supports three reasoning modes through the thinking parameter:
- `enabled` — Reasoning is always enabled.
- `adaptive` — M3 automatically determines when additional reasoning is beneficial.
- `disabled` — Reasoning is disabled to minimize latency and maximize throughput.
Local Deployment
Download the model:
hf download MiniMaxAI/MiniMax-M3 --local-dir MiniMax-M3We recommend the following inference frameworks to serve the model:
- SGLang - see SGLang cookbook.
- vLLM - see vLLM recipes.
- Transformers - see Transformers docs.
Inference Parameters
We recommend the following parameters for best performance: temperature=1.0, top_p=0.95.
Contact Us
Contact us at model@minimax.io.