benchgap
Moonshot · model

Kimi K2 benchmark scores

As of 2026-11-21, Kimi K2 (Moonshot) has measured scores on 21 benchmarks and estimated scores on 48 more.

BenchmarkScoreSource
AA-GPQA Diamond76.6%measured
AA-HLE7.4%measured
AA-SciCode41.6%estimated ± 3.8 pp, high confidence
AIME2652.5%estimated ± 4.3 pp, medium confidence
Artificial Analysis Intelligence Index12.7%measured
CritPt0.0%measured
GDPval-AA3.3%estimated ± 11.0 pp, low confidence
GPQA Diamond75.3%estimated ± 1.2 pp, high confidence
HLE4.7%measured
IFBench54.3%estimated ± 6.6 pp, low confidence
AA-LCR53.0%measured
LiveCodeBench v653.7%measured
AA-Omniscience Accuracy27.4%measured
SciCode40.9%estimated ± 3.6 pp, high confidence
BrowseComp57.5%estimated ± 13.0 pp, low confidence
IFEval89.8%measured
LongBench v252.7%estimated ± 5.5 pp, medium confidence
τ²-bench results61.1%measured
AA Coding Index33.8%estimated ± 5.3 pp, medium confidence
APEX-Agents-AA1.6%estimated ± 11.4 pp, low confidence
BioMysteryBench (human-difficult)0.0%estimated ± 6.8 pp, low confidence
SWE-bench Verified65.8%measured
BFCL v450.6%estimated ± 8.5 pp, low confidence
SWE-bench Pro42.8%estimated ± 4.5 pp, high confidence
AA Agentic Index12.8%estimated ± 12.7 pp, low confidence
MMLU-Pro81.1%measured
AA-IFBench41.5%measured
AIME 202469.6%measured
BBH95.1%estimated ± 2.4 pp, low confidence
GPQA75.1%measured
MMLU89.5%measured
Vals GPQA Diamond75.7%estimated ± 3.2 pp, high confidence
Vals LiveCodeBench65.9%estimated ± 10.0 pp, low confidence
Vals MMLU-Pro81.1%estimated ± 2.3 pp, high confidence
MATH-50097.4%measured
MMMLU83.3%estimated ± 0.7 pp, medium confidence
AIME 202549.5%measured
AA MMLU-Pro85.2%estimated ± 0.8 pp, high confidence
ARC-AGI-135.9%estimated ± 6.7 pp, low confidence
ARC-AGI-20.0%estimated ± 10.9 pp, low confidence
FrontierCode 1.1 Main2.4%estimated ± 4.2 pp, low confidence
Vals SWE-bench57.1%estimated ± 6.6 pp, medium confidence
HealthBench Professional (length-adjusted, Artificial Analysis)49.7%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)46.6%estimated ± 1.1 pp, low confidence
DeepSWE22.0%estimated ± 12.7 pp, low confidence
HLE w/o tools26.0%estimated ± 4.9 pp, high confidence
SWE Multilingual47.3%measured
ARC-AGI-31.3%estimated ± 9.8 pp, low confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
Vibe Code Bench0.0%estimated ± 8.4 pp, medium confidence
C-Eval89.4%estimated ± 1.0 pp, medium confidence
HMMT Feb 202626.4%estimated ± 11.0 pp, low confidence
MMLU-Redux91.0%estimated ± 0.6 pp, high confidence
NL2Repo19.3%estimated ± 6.2 pp, low confidence
SuperGPQA57.2%measured
HealthBench Hard0.0%estimated ± 8.6 pp, low confidence
MMLU-Pro (Arcee)77.1%estimated ± 4.0 pp, low confidence
React Native Evals73.3%estimated ± 2.7 pp, medium confidence
SWE-Rebench57.0%estimated ± 3.6 pp, low confidence
BioMysteryBench (human-solvable)81.7%estimated ± 0.6 pp, low confidence
HealthBench (length-adjusted)54.1%estimated ± 3.7 pp, medium confidence
HealthBench Professional56.7%estimated ± 4.2 pp, low confidence
HLE-Verified22.1%estimated ± 7.1 pp, low confidence
STEMBench2.4%estimated ± 0.2 pp, medium confidence
SimpleQA31.0%measured
LiveCodeBench70.1%estimated ± 14.6 pp, low confidence
MRCRv223.1%estimated ± 11.3 pp, low confidence
MMLongBench-Doc59.7%estimated ± 2.6 pp, low confidence
BioBasicsBench34.1%estimated ± 3.8 pp, low confidence