MiniMax M3.1 Flash: Redefining Agent Performance and Cost - MiniMax Research
Flat pricing is set at $0.15 per 1M input tokens, $0.60 per 1M output tokens, and $0.015 per 1M cached input tokens.
- Output speeds reach over 200 tokens per second with a median P50 speed of 170+ TPS.
- Available immediately through the MiniMax API, M Plan, and MiniMax Code, with model weights scheduled for a later release.
- Scores 73.7% on DeepSWE v1.1, 61.7 points on OSWorld V2, and 82% on CyberGym.
- Reduces overall usage costs for agentic tasks by 70% compared to its predecessor.
Developers building coding and agentic computer-use tools can access frontier-level benchmark capabilities at higher inference speeds and lower token costs.
