Unverified50% confidenceBenchmarkExact time
MiniMax released M3.1 Flash on October 19, scoring 73.7% on DeepSWE v1.1, 76.2 on BankerToolBench, 61.7 on OSWorld V2, 29.6 on Agents' Last Exam, and 82% on CyberGym; it ranks first among Chinese models on the OSWorld V2 and ALE benchmarks.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/20/2026
First Seen
Valid until: 11/3/2026
Sources
Related Entities
Related Claims
UnverifiedOn October 19, MiniMax M2.5/M2.7 were closed to new users and plan requests were routed to M3.1 Flash; on the same day `minimax-m3.1-flash` launched on AIHubMix on a pay-per-token basis.75% similarPartially VerifiedMaximum output tokens: MiniMax M3.1 Flash is 512K, DeepSeek V4.1 Flash 1017 is 384K, and Claude Haiku 5.5 is 128K.73% similarUnverifiedOn AIHubMix, MiniMax M3.1 Flash's input price is about 44% lower than M2.7's ($0.165 vs. $0.2958) and about 43% lower than M3's ($0.165 vs. $0.288), with the same ratios on the output side.70% similarUnverifiedOn the DeepSWE coding leaderboard, Beam scored 44.4 points, ranking last, below DeepSeek V4.1 Flash's 74.2 points, Kimi K3's approximately 68 points, GLM's approximately 61 points, and Qwen3 Max's approximately 51 points.69% similarUnverifiedMiniMax M3.1 Flash's thinking tokens count toward the max_tokens limit, so reusing the 4K or 8K caps from the M2.x era may cause truncation with finish_reason: "length"; MiniMax recommends raising max_tokens to 131,072, with an upper limit of 524,288.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1074133API
curl https://kongchang.com/api/v1/knowledge/claims/1074133MCP
get_claim(id=1074133)