Kimi K2.6 — Benchmarks

Benchmark scores for Kimi K2.6 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #8 of 73 open modelscomposite 68.5/100 across 7 benchmarks in 4 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
LiveBench Coding78.6#2 / 8#15 / 37

Knowledge

BenchmarkScoreOpen rankAll models
SimpleQA38.7%#4 / 11#35 / 65

Math

BenchmarkScoreOpen rankAll models
AIME 2024/202596.1%#3 / 34#14 / 155
LiveBench Math84.3#4 / 8#30 / 37
FrontierMath39.0%#1 / 12#15 / 101

Reasoning

BenchmarkScoreOpen rankAll models
GPQA Diamond90.8%#1 / 46#17 / 182
LiveBench Reasoning79.4#3 / 8#27 / 37

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