Qwen3.8 27B — Benchmarks

Benchmark scores for Qwen3.8 27B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #53 of 123 open modelscomposite 51.3/100 across 6 benchmarks in 3 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
SciCode44.7%#18 / 50#88 / 155
LiveBench Coding75.7#7 / 16#36 / 54

Math

BenchmarkScoreOpen rankAll models
LiveBench Math86.2#8 / 16#38 / 54
ProofBench16.0%#12 / 21#43 / 64

Reasoning

BenchmarkScoreOpen rankAll models
LiveBench Reasoning80.0#8 / 16#39 / 54
CritPt5.4%#12 / 50#78 / 167

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