moonshotai/Kimi-K3
ale-cli-pass-rate · 27.61904761904762 source-defined score; unit unspecified
Source-qualified evaluations, report statistics and safety observations.
ale-cli-pass-rate · 27.61904761904762 source-defined score; unit unspecified
de · 85.5 source-defined score; unit unspecified
SWE_Bench_Pro · 59.5 source-defined score; unit unspecified
table · 65.45 source-defined score; unit unspecified
diamond · 92.8 source-defined score; unit unspecified
deep_swe · 77 source-defined score; unit unspecified
ko · 14.7 source-defined score; unit unspecified
swe_bench_%_resolved · 72 source-defined score; unit unspecified
MathArena/aime_2026 · 92.6 source-defined score; unit unspecified
reasoning · 38.73 source-defined score; unit unspecified
fr · 57.3 source-defined score; unit unspecified
SWE_Bench_Pro · 42.9 source-defined score; unit unspecified
accuracy · 0.743 source-defined score; unit unspecified
swe_atlas_qna · 49.5 source-defined score; unit unspecified
hle · 14.9 source-defined score; unit unspecified
swe_bench_%_resolved · 53.73 source-defined score; unit unspecified
terminalbench_2 · 63.5 source-defined score; unit unspecified
en · 86.2 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.