minimax-m3.1POST
MiniMax M3.1 Flash launch: nearly 3x faster inference and 70% lower cost
A summary of the official launch: M3.1 Flash targets agent, coding, computer-use and security work, with nearly 3x faster inference, over 200 TPS, DeepSWE v1.1 at 73.7% and OSWorld V2 at 61.7.
渣渣渣渣渣
@huang18825Archived 8 days agoOriginally posted 2026.10.19
渣渣渣渣渣 @huang18825Originally posted 2026.10.19
MiniMax 发布 M3.1 Flash，推理速度提升近 3 倍，成本降低 70% MiniMax 正式发布 M3.1 Flash，主打 Agent、编程、电脑操作及网络安全能力。官方称，该模型线上推理速度较前代提升近 3 倍，日常场景输出速度超过 200 TPS，在 DeepSWE v1.1 测试中得分达 73.7%，OSWorld V2 得分为 61.7。
x.com/huang18825/status/2112150770195767518Archived 2026.10.20 20:26