AI Models & Agents
MiniMax M3.1 Flash targets agents and coding
The company says its new model runs faster and costs less, but its performance figures have not been independently validated.
Editorial illustration of an AI model assisting with code, computer tasks and automated workflows.
Photo: Kronos News
MiniMax announced M3.1 Flash on October 19, positioning the model for agentic tasks, coding, computer use and cybersecurity. The company says inference is nearly three times faster and agentic-task usage costs are 70% lower than with its predecessor. [1][2]
MiniMax reported scores including 73.7% on DeepSWE v1.1 and 61.7 on OSWorld V2; these are company-published benchmark results, not evidence of performance in everyday deployments. SiliconANGLE noted that the DeepSWE results it compared had not been independently checked. [1][2] The model is available through MiniMax’s API, M Plan subscription and MiniMax Code, while its weights are planned for later release. [1][2]
Editorial notes
Transparency note
AI assisted drafting. Human edited and reviewed.
- AI assisted
- Yes
- Human review
- Yes
- Last updated
Risk assessment
Performance, speed and cost figures are company claims; benchmark results should not be treated as independently validated or proof of real-world effectiveness.
Sources
Related stories
View all
Topics
About the author
Kronos News Desk covers ai models & agents and editorial analysis for Kronos News.