跳到正文
RadarAI· X:Testing Catalog (@testingcatalog)·· 3 小时前同新闻AI 评分78

MiniMax 开源 MiniMax M3.1 Flash 权重,428B 总参数每 token 激活 23B

AI 导读

MiniMax 在 HuggingFace 和 ModelScope 公开 MiniMax M3.1 Flash 权重,总参数 428B,每 token 激活 23B。

同一新闻,精选展示《MiniMax 开源 M3.1 Flash 模型权重,总参数 428B》

正文 · 原文

🚨 AI News | TestingCatalog· @testingcatalog · X·2026-10-23 17:18· 19 分钟前AI 评分73

AI 导读

MiniMax 在 HuggingFace 和 ModelScope 公开 MiniMax M3.1 Flash 权重,总参数 428B,每 token 激活 23B。该模型即 OpenRouter 上的 "Space Bunny" 隐身模型,10 月 19 日公布,现已登陆 MiniMax 平台和 API;官方给出 BankerToolBench 76.2、DeepSWE v1.1 73.7%、OSWorld V2 61.7、Agents' Last Exam 29.6、CyberGym 82.2 等智能体、编码与安全成绩,并采用 MiniMax Sparse Attention 和 DSpark 投机解码提升推理效率。

正文 · 原文

MiniMax has published MiniMax M3.1 Flash weights on HuggingFace.

The model has 428B parameters total, 23B activated per token.

MiniMax M3.1 Flash, also known as "Space Bunny" stealth model from OpenRouter, was announced on October 19 and is now available on MiniMax platform and APIs.

引用MiniMax (official)@MiniMax_AI
MiniMax M3.1 Flash weights are now public. 428B parameters total, 23B activated per token. Agentic performance: A generational leap in agentic tasks, coding, and cybersecurity — 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. Efficient by design: MiniMax Sparse Attention and DSpark speculative decoding improve inference efficiency. Hugging Face: https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash ModelScope: https://modelscope.cn/models/MiniMax/MiniMax-M3.1-Flash API: http://platform.minimax.io M Plan: https://minimax.io/m-planMiniMax MCode for desktop: https://agent.minimax.io/download
在 X 查看被引用的帖子

来源:🚨 AI News | TestingCatalog · x.com

来源:RadarAI · aihot.news