MiniMax M3开源了。

一般愿意开源的，都是过时的，还有未开源的弹药未释放。

Hugging Face 主仓库（推荐）：

https://huggingface.co/MiniMaxAI/MiniMax-M3… 这里提供完整模型权重（~428B 总参数、~23B 激活参数，原生多模态，支持 1M 上下文），支持 Safetensors/PyTorch 格式下载，还有量化版本（如 MiniMaxAI/MiniMax-M3-MXFP8）。

GitHub 仓库（配套代码和文档）： https://github.com/MiniMax-AI/MiniMax-M3… 包含推理指南、MSA（MiniMax Sparse Attention）相关支持等。

MiniMaxAI/MiniMax-M3 · Hugging Face

Source: https://huggingface.co/MiniMaxAI/MiniMax-M3

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.

https://huggingface.co/MiniMaxAI/MiniMax-M3#minimax-sparse-attention-msaMiniMax Sparse Attention (MSA)

M3 is powered byMiniMax 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.

📄 Read the technical report:arXiv:2606.13392·Hugging Face Papers

https://huggingface.co/MiniMaxAI/MiniMax-M3#how-to-useHow to Use

• MiniMax Agent
• MiniMax API

M3 supports three reasoning modes through thethinkingparameter:

• enabled— Reasoning is always enabled.
• adaptive— M3 automatically determines when additional reasoning is beneficial.
• disabled— Reasoning is disabled to minimize latency and maximize throughput.

https://huggingface.co/MiniMaxAI/MiniMax-M3#local-deploymentLocal Deployment

Download the model:

hf download MiniMaxAI/MiniMax-M3 --local-dir MiniMax-M3

We recommend the following inference frameworks to serve the model:

• SGLang- seeSGLang cookbook.
• vLLM- seevLLM recipes.
• Transformers- seeTransformers docs.
• KTransformers- seeKTransformers MiniMax-M3 tutorial.
• unsloth- seetutorial

https://huggingface.co/MiniMaxAI/MiniMax-M3#inference-parametersInference Parameters

We recommend the following parameters for best performance:temperature=1\.0,top\_p=0\.95.

https://huggingface.co/MiniMaxAI/MiniMax-M3#contact-usContact Us

Contact us at[email protected].