Qwen3.6 27B — Benchmarks

Benchmark scores for Qwen3.6 27B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #16 of 78 open modelscomposite 56/100 across 11 benchmarks in 3 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
LiveBench Coding71.8#13 / 19#52 / 63
SciCode37.3%#32 / 62#126 / 161

Math

BenchmarkScoreOpen rankAll models
FrontierMath35.1%#12 / 15#54 / 102
LiveBench Math79.9#16 / 19#56 / 63
AIME 2024/202591.1%#10 / 82#58 / 270

Reasoning

BenchmarkScoreOpen rankAll models
Chess Puzzles22.0%#6 / 65#74 / 205
Mystery Game Puzzles7.0%#15 / 17#111 / 122
DTBench78.1%#26 / 67#119 / 210
LMCA34.5%#15 / 53#104 / 172
GPQA Diamond85.9%#17 / 94#75 / 289
CritPt0.9%#34 / 62#119 / 172
LiveBench Reasoning70.3#19 / 19#62 / 63

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