Frontier Log

MiniMaxModel

MiniMax-M1

Made by
MiniMax
Released
Announced in
MiniMax-M1, the World's First Open-Source, Large-Scale, Hybrid-Attention Reasoning Modelminimax.io
Event
MiniMax-M1
API record
/api/v1/models/minimax-m1

Benchmark results

BenchmarkScoreSource check
Humanity's Last ExamNo tools · MiniMax-M1-80k, 80K thinking budget; text-only subset8.4%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
SWE-bench VerifiedMiniMax-M1-80k, 80K thinking budget56.0%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
CorroboratedThe same number is printed in 1 other lab document: 1
GPQA DiamondMiniMax-M1-80k, 80K thinking budget70.0%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
AIME 2024No tools · MiniMax-M1-80k, 80K thinking budget86.0%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
AIME 2025No tools · MiniMax-M1-80k, 80K thinking budget76.9%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
MMLU-ProMiniMax-M1-80k, 80K thinking budget81.1%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
LiveCodeBenchMiniMax-M1-80k, 80K thinking budget; problems from Aug 2024 to May 202565.0%
CheckedA person checked this number against its source on 19 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced.
Sources

Reported by MiniMax

Events

All models