Refresh Models —Deals —Regulatory —Sources —

Qwen3.5 27B vs Qwen3.5 2B

35 SHARED BENCHMARKS

Across 35 shared benchmarks, Qwen3.5 27B scores higher on 34 and Qwen3.5 2B on 1. The widest gap is HLE, where Qwen3.5 27B scores 48.5 against 5.

ALIBABAVSALIBABA35 SHARED34–1 HEAD-TO-HEAD

BenchmarkQwen3.5 27BMARGINQwen3.5 2B

AA Agentic Index54.60.8

AA Intelligence22.96.9

AA-LCR77.725.6

AA-Omniscience-44-60.2

Artificial Analysis Coding Index34.92.9

bfcl_v468.543.6

C-Eval90.573.2

GPQA Diamond85.851.6

HallusionBench7065.5

HLE48.55

IFBench76.541.3

ifeval9578.6

include81.655.4

longbench_v260.638.7

MathVista87.848.7

mmlu_prox82.252.3

mmlu_redux93.279.6

MMLU-Pro86.166.5

mmmlu85.963.1

MMStar8155.1

multichallenge60.833.7

OCRBench89.486.1

OmniScience Accuracy20.78.1

OmniScience Non-Hallucination24.925.8

RealWorldQA83.765.1

scicode39.57.2

SimpleVQA5635.2

supergpqa65.637.5

Terminal-Bench Hard32.63.8

τ²-Bench Telecom (AA run)93.981.6

τ³-Bench68.46.2

Best tracked score per model per benchmark (default configuration; source-attributed). ↓ marks lower-is-better metrics. Open either model for its full surface, provenance and pricing.

Compare Qwen3.5 27B withALL PAIRINGS →

AlibabaQwen3.5 35B A3B Qwen3 VL 4B (Reasoning)Qwen3.5 397B A17B Qwen2.5 Instruct 72B Qwen3.6 Plus Qwen3.6 35B A3B

Across the fieldQwen3.6 27B Qwen3.7 Max Qwen3.8 Max Preview Qwen3.7 Plus Preview Claude Opus 4 Qwen3.5 4B Gemini 3.6 Flash Nemotron 3 Super 120B A12B

Compare Qwen3.5 2B withALL PAIRINGS →

AlibabaQwen3.5 35B A3B Qwen3 VL 4B (Reasoning)Qwen3.5 397B A17B Qwen2.5 Instruct 72B Qwen3.6 Plus Qwen3.6 35B A3B

Across the fieldClaude Sonnet 5 Muse Spark 1.1 GLM-5.1 GLM-5.2 o4-mini Claude Sonnet 4.6 o1 Gemma 4 31B