MiniMax M3.1 Flash
MiniMax-M3.1-Flash is a multimodal model from MiniMax. Compared with M3, it delivers significant improvements across multiple benchmarks, particularly in complex agentic tasks, coding, and cybersecurity.
Specifications
| Lab | MiniMax |
|---|---|
| Context window | 1M tokens |
| Input price | $0.00/1M |
| Output price | $0.00/1M |
| Release | Oct 2026 |
Strengths
- Improved agentic, coding and cybersecurity performance over MiniMax M3
- 1M-token context window with image and video input
- Adjustable reasoning effort (low through max)
- Free tier on OpenRouter for low-cost experimentation
Weaknesses
- Brand-new release with limited independent benchmark coverage
- Available only as a free-tier listing so far, with availability limits
- Smaller ecosystem and tooling than Western frontier labs
Best for
- Agentic coding and multi-step autonomous tasks
- Long-context multimodal analysis over documents and video
- Cost-sensitive prototyping on the free tier
In Depth: MiniMax M3.1 Flash
Summary
MiniMax M3.1 Flash is an AI model from MiniMax.
Released in Oct 2026. It takes text, image, and video input and produces text output, with a context window of 1M tokens. Input pricing is $0.00 per 1M tokens and output is $0.00 per 1M tokens on OpenRouter.