Qwen3.5 35B A3B — Benchmarks

Benchmark scores for Qwen3.5 35B A3B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #21 of 78 open modelscomposite 54.7/100 across 8 benchmarks in 4 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
SciCode29.3%#49 / 62#144 / 161
LMArena WebDev1250.3#44 / 48#110 / 123

Instruction Following

BenchmarkScoreOpen rankAll models
LMArena Text1395.0#50 / 182#149 / 391

Knowledge

BenchmarkScoreOpen rankAll models
MMLU-Pro85.3%#7 / 144#32 / 257

Math

BenchmarkScoreOpen rankAll models
AIME 2024/202570.0%#27 / 82#130 / 270

Reasoning

BenchmarkScoreOpen rankAll models
Chess Puzzles10.0%#24 / 65#131 / 205
DTBench80.0%#23 / 67#114 / 210
LMCA29.5%#22 / 53#115 / 172
GPQA Diamond83.5%#21 / 94#89 / 289
CritPt0.6%#36 / 62#123 / 172

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