MiniMaxAI/MiniMax-M2.7
skillsbench_v1_1 · 34.9 source-defined score; unit unspecified
Source-qualified evaluations, report statistics and safety observations.
skillsbench_v1_1 · 34.9 source-defined score; unit unspecified
terminalbench_4 · 38.1 source-defined score; unit unspecified
terminalbenchscience_0_1 · 0.5 source-defined score; unit unspecified
es · 75.7 source-defined score; unit unspecified
text_formatting · 21.52 source-defined score; unit unspecified
MathArena/aime_2026 · 69.1 source-defined score; unit unspecified
diamond · 74.2 source-defined score; unit unspecified
deep_swe · 28 source-defined score; unit unspecified
swe_bench_%_resolved · 60.47 source-defined score; unit unspecified
long · 37.74 source-defined score; unit unspecified
diamond · 81.8181818182 source-defined score; unit unspecified
swe_bench_multilingual_%_resolved · 78.9 source-defined score; unit unspecified
th · 75 source-defined score; unit unspecified
chart · 64.7 source-defined score; unit unspecified
overall · 56.3 source-defined score; unit unspecified
mmlu_pro · 30.43 source-defined score; unit unspecified
gsm8k · 81.73 source-defined score; unit unspecified
diamond · 90.5 source-defined score; unit unspecified
These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.