Key Info

MiniMax released the public weights of M3.1 Flash, a 428B-parameter model with 23B parameters activated per token, available on Hugging Face and ModelScope alongside API access.

Highlights

  • Agentic benchmarks: 76.2 on BankerToolBench, 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam, 82.2 on CyberGym
  • Uses MiniMax Sparse Attention and DSpark speculative decoding to improve inference efficiency
  • Weights published on Hugging Face and ModelScope; also offered via API, M Plan, and MCode for desktop
  • MiniMax says Space Bunny, which topped global leaderboards with 200T+ tokens, was an earlier version of M3.1 Flash