MiniMax Open-Sources M3.1 Flash: 428B Params, 23B Active, NVIDIA Vera Rubin Day-0 Support
NVIDIAAI · x · 2026-10-23
MiniMax has released and open-sourced M3.1 Flash weights: 428B total parameters with 23B activated per token, targeting agentic tasks, coding, and cybersecurity. Claimed scores: 76.2 on BankerToolBench, 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam, and 82.2 on CyberGym. Efficiency comes from MiniMax Sparse Attention and DSpark speculative decoding.
NVIDIA announced its first-ever Day 0 support: the model runs out-of-the-box on Vera Rubin with up to 200K TPS/GPU throughput and over 700 TPS per user (SemiAnalysis AgentX, NVFP4 + spec decoding). Weights are on Hugging Face and ModelScope, with API, M Plan, and desktop MCode available.
Related event: MiniMax Open-Sources M3.1-Flash: 428B Multimodal MoE with 1M Context(8 posts)→
More from Infra
- NVIDIA's $1.2M HGX Rubin NVL8 hits GTC Berlin, promising 10x tokens per megawatt over B200 — thursdai_pod · 2026-10-23
- Amazon scraps plans for four data centers on $427M former GWU Ashburn campus — GaryMarcus · 2026-10-23
- Developer platforms raided for AI inference keys: Railway, Fireworks, LiteLLM all hit this month — ccerrato147 · 2026-10-23
- Koi Paper: Self-Calibrating Planner Boosts Heterogeneous LLM Fleet Goodput 1.18x, Cuts $/token 20% — burny_tech · 2026-10-23
- Inco Open-Sources DFlash 2 Drafter: Lossless 3.5× Speedup for MiniMax M3.1 Flash — Lianhuiq · 2026-10-23
- Cloudflare CEO: agent traffic will be 1000x human traffic in 5 years, internet needs micropayments — 4KTV · 2026-10-23