Kimi K2.5 — Benchmarks

Benchmark scores for Kimi K2.5 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #19 of 73 open modelscomposite 57.1/100 across 10 benchmarks in 4 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
Terminal-Bench43.2%#3 / 16#30 / 57
SWE-bench Multilingual67.3%#3 / 4#7 / 14

Knowledge

BenchmarkScoreOpen rankAll models
Humanity's Last Exam24.4%#1 / 4#13 / 46
SimpleQA33.9%#6 / 11#44 / 65
MMLU-Pro87.1%#3 / 119#17 / 259

Math

BenchmarkScoreOpen rankAll models
AIME 2024/202592.2%#5 / 34#28 / 155
FrontierMath27.9%#3 / 12#27 / 101

Reasoning

BenchmarkScoreOpen rankAll models
GPQA Diamond87.6%#5 / 46#33 / 182
ARC-AGI65.3%#1 / 10#55 / 158
SimpleBench46.8%#7 / 19#47 / 90

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