MiniMax: MiniMax-M3 model card
Full side-by-side comparison of both captured documents.
Both captured versions of this document, in full. Deleted text is struck through on the left; inserted text is underlined on the right.
direct captureBefore
Captured 2026-07-20 05:13:34 UTCMethod directHash 1ee242b6c8a2ecaddirect captureAfter
Captured 2026-07-23 04:27:21 UTCMethod directHash 5965f13cd579f61braw before · raw after · provider’s live document
| Before | After |
|---|---|
--- pipeline_tag: image-text-to-text license: other license_name: minimax-community license_link: LICENSE library_name: transformers tags: - multimodal - moe - agent - coding - video --- 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. ## MiniMax Sparse Attention (MSA) M3 is powered by [**MiniMax Sparse Attention (MSA)**](https://github.com/MiniMax-AI/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](https://arxiv.org/abs/2606.13392) · [Hugging Face Papers](https://huggingface.co/papers/2606.13392) ## How to Use - [MiniMax Agent](https://agent.minimax.io/) - [MiniMax API](https://platform.minimax.io/) 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: ```bash hf download MiniMaxAI/MiniMax-M3 --local-dir MiniMax-M3 ``` We recommend the following inference frameworks to serve the model: - [SGLang](https://docs.sglang.io/) - see [SGLang cookbook](https://docs.sglang.io/cookbook/autoregressive/MiniMax/MiniMax-M3). - [vLLM](https://github.com/vllm-project/vllm) - see [vLLM recipes](https://recipes.vllm.ai/MiniMaxAI/MiniMax-M3). | |
- [Transformers](https://github.com/huggingface/transformers) - see [Transformers docs](https://huggingface.co/docs/transformers/model_doc/minimax_m3_vl). - [KTransformers](https://github.com/kvcache-ai/ktransformers) - see [KTransformers MiniMax-M3 tutorial](https://github.com/kvcache-ai/ktransformers/blob/main/doc/en/kt-kernel/MiniMax-M3-Tutorial.md). - [unsloth](https://unsloth.ai) - see [tutorial](https://unsloth.ai/docs/models/minimax-m3) | - [Transformers](https://github.com/huggingface/transformers) - see [Transformers docs](https://huggingface.co/docs/transformers/model_doc/minimax_m3_vl). - [KTransformers](https://github.com/kvcache-ai/ktransformers) - see [KTransformers MiniMax-M3 tutorial](https://github.com/kvcache-ai/ktransformers/blob/main/doc/en/kt-kernel/MiniMax-M3-Tutorial.md). - [unsloth](https://unsloth.ai) - see [tutorial](https://unsloth.ai/docs/models/minimax-m3) |
- [ATOM](https://github.com/ROCm/ATOM/tree/main) - see [MiniMax-M3 MXFP4/MXFP8 Usage Guide](https://github.com/ROCm/ATOM/blob/main/recipes/MiniMax-M3.md) | |
### 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](mailto:model@minimax.io). | ### 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](mailto:model@minimax.io). |