Breaking
MiniMax launches M3.1 Flash at a third of M3's cost and three times the speed
First seen on X 14 hours ago@MiniMax_AI ♥ 3,9827 views
Chinese AI company MiniMax has released M3.1 Flash, a lightweight model built for agentic work, and made it available through its API, its M Plan subscription and the MiniMax Code app.
MiniMax said on its blog and on X on Oct 19 that M3.1 Flash cuts the cost of agentic tasks by 70% compared with M3 and runs nearly three times faster, with output speeds above 200 tokens per second. API pricing is a flat $0.15 per million input tokens, $0.60 per million output tokens and $0.015 for cached input. The company reported 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2 and 82% on CyberGym. It also said Space Bunny, an anonymous model that processed more than 200 trillion tokens, was an early version of M3.1 Flash. Weights will be released on Oct 23. The announcement passed 510,000 views and 3,900 likes in about two hours. One user replied that "$0.60 per million tokens just beat Opus 5." MiniMax said a larger Pro model will follow.
“MiniMax M3.1 Flash: Space Performance + Bunny Speed 🚀🐰 Yes, Space Bunny - which has topped global leaderboards with 200T+ tokens - was an earlier version of M3.1 Flash.”
