Model record

Kimi K2 Thinking

Open weights

Moonshot AI · released Nov 6, 2025

Overall standing

Overall
44.5rank 45 of 63
44.5 is at approximately the 29th percentile of the ranked Overall field. The field spans 19.0 to 62.2, with a median of 49.4.
6 of 8 measured · 2 estimated
Category standingsReasoning 60.2Coding 48.7Math 2.6Agentic 15.7
Reasoning
60.2rank 44 of 63
60.2 is at approximately the 31th percentile of the ranked Reasoning field. The field spans 31.6 to 73.3, with a median of 65.0.
4 of 4 measured
Coding
48.7rank 41 of 63
48.7 is at approximately the 36th percentile of the ranked Coding field. The field spans 10.4 to 72.4, with a median of 54.5.
1 of 2 measured · 1 estimated
Math
2.6rank 43 of 63
2.6 is at approximately the 32th percentile of the ranked Math field. The field spans 0.0 to 32.3, with a median of 5.7.
1 of 1 measured
Agentic
15.7rank 41 of 63
15.7 is at approximately the 36th percentile of the ranked Agentic field. The field spans 3.3 to 33.4, with a median of 19.0.
0 of 1 measured · 1 estimated
Model factsOpen weights · Price not listed

Model facts come from the provider page. Listed API prices carry a separate first-party source and check date.

Scores and sources

Measured scores retain their source and retrieval date. Missing scores are not treated as zero.

Benchmark scores, evaluation settings, sources, and retrieval dates for Kimi K2 Thinking
BenchmarkCategoryScoreEvaluation settingsSource
Terminal-Bench v2.1CodingNot measured in this datasetNo source
τ³-BankingAgenticNot measured in this datasetNo source
AA-LCRReasoning66.3Artificial Analysis independent run; 100 roughly 100k-token multi-document questions; 3 repeats; equality-checker pass@1; no toolsOpen source (opens in a new tab)
Humanity's Last ExamReasoning22.3Artificial Analysis independent run; 2,158 text-only questions; pass@1; GPT-4o (August 2024) equality checker; no toolsOpen source (opens in a new tab)
GPQA DiamondReasoning83.8Artificial Analysis independent run; 198 questions; 5 repeats; regex-graded pass@1; no toolsOpen source (opens in a new tab)
SciCodeCoding42.4Artificial Analysis independent run; 288 test subproblems; scientist background included; 3 repeats; subproblem pass@1Open source (opens in a new tab)
IFBenchReasoning68.1Artificial Analysis independent run; 294 single-turn prompts; 5 repeats; official loose evaluator; prompt-level pass@1Open source (opens in a new tab)
CritPtMath2.6Artificial Analysis independent run; 70 test challenges; 5 repeats; two-step answer parsing; official CritPt grader; pass@1; no toolsOpen source (opens in a new tab)