# MiniMax releases M3.1-Flash, a 428B MoE model for coding and agents

> Original title: 1M tokens, native image and video input, and 23B active parameters per token. MiniMax-M3.1-Flash has landed as a 428B MoE for coding, com...

- Source: ModelScope
- Published: 2026-10-23T02:08:54.000Z
- Bad Influence: https://badinfluence.news/items/mkur47rp14sgr7lmhtaktmup6
- Original: https://x.com/ModelScope2022/status/2113452574292550076

## Summary

MiniMax-M3.1-Flash is a 428B mixture-of-experts model with a 1M-token context window, native image and video input, and 23B active parameters per token. ModelScope says it targets coding, computer use, and tool-driven agents, scoring 62.8 on Automation-Bench v1.0.6 and 73.7% on DeepSWE v1.1. Model weights are available with SGLang and vLLM deployment guidance.

## Post

1M tokens, native image and video input, and 23B active parameters per token. MiniMax-M3.1-Flash has landed as a 428B MoE for coding, computer use, and tool-driven agents.
🤖 https://modelscope.ai/models/MiniMax/MiniMax-M3.1-Flash
📄 https://modelscope.ai/papers/2606.13392  

🏆 Top result for agent execution: Scores 62.8 on Automation-Bench v1.0.6, the highest result in the model card’s reported comparison.
💻 Strong at coding and computer use: Reaches 73.7% on DeepSWE v1.1, 61.7 on OSWorld 2.0, and 29.6 on Agents’ Last Exam.
🔐 Built for cybersecurity work: Scores 82.2 on CyberGym for vulnerability discovery and validation.
⚡ Efficiency is built in: Sparse attention throughout, QAT NVFP4, and the included DSpark draft head support efficient inference and speculative decoding.
🛠️ Ready to self-host: Model weights are available with SGLang and vLLM deployment guidance.

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHVR96oBWgAArjee.jpg&mode=thumb&exp=1793404800&sig=e432f73f2ee46a86)