MiniMax M2.7 — Benchmarks

Benchmark scores for MiniMax M2.7 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #119 of 123 open modelscomposite 16.5/100 across 4 benchmarks in 3 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
Terminal-Bench45.1%#2 / 16#28 / 57
SciCode47.0%#12 / 50#76 / 155

Math

BenchmarkScoreOpen rankAll models
ProofBench3.0%#18 / 21#60 / 64

Reasoning

BenchmarkScoreOpen rankAll models
CritPt0.6%#31 / 50#117 / 167

Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.