The First Published NVFP4 Quantization of MiniMax M3

NOTICE

This is an experimental quantization of MiniMax M3 to NVFP4 for use on DGX Spark. (Note: This quantization is not DGX Spark only.)

For DGX Spark Users: Run with sparkrun; part of Spark Arena

https://sparkrun.dev https://spark-arena.com

To run with sparkrun on 4x DGX Spark Nodes:

sparkrun run @experimental/minimax-m3-v0-nvfp4-4x

For RTX Pro 6000 Users (or DGX Spark Users who don't want to use sparkrun): You can run this using the custom sglang container:

docker pull scitrera/dgx-spark-sglang-mm:v0

(Container build is multi-arch so it can be used for x86 and ARM)

Reference settings can be derived from the sparkrun recipe: https://github.com/spark-arena/recipe-registry/blob/main/experimental-recipes/minimax-m3/minimax-m3-v0-nvfp4-4x.yaml

Happy Coding! Let's go!

---

---

👁 MiniMax Agent
👁 API
👁 MiniMax Website

👁 WeChat
👁 Discord
👁 Hugging Face
👁 GitHub
👁 arXiv Paper
👁 LICENSE

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.

👁 Image

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.

👁 GQA vs MSA Efficiency Comparison

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

How to Use

• MiniMax Agent
• MiniMax API

M3 supports two reasoning modes:

• thinking — for complex reasoning, agentic tasks, and long-horizon collaboration.
• non-thinking — for latency-sensitive scenarios such as chat and code completion.

Local Deployment

Download the model:

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

We recommend the following inference frameworks (listed alphabetically) 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, top_k=40.

Contact Us

Contact us at model@minimax.io.

Base model

MiniMaxAI/MiniMax-M3

(21)

3 items • Updated 5 days ago • 6