VECTOR WIREAI INTELLIGENCE
⌘K
PKT
Refresh Models —Deals —Regulatory —Sources —

Questions this page answers

Which is better, Qwen2.5 VL 32B Instruct or Qwen2 VL 72B Instruct?
Across 11 shared benchmarks, Qwen2.5 VL 32B Instruct scores higher on 9 and Qwen2 VL 72B Instruct on 2. The widest gap is mathvision, where Qwen2.5 VL 32B Instruct scores 40 against 25.9.

Qwen2.5 VL 32B Instruct vs Qwen2 VL 72B Instruct

Across 11 shared benchmarks, Qwen2.5 VL 32B Instruct scores higher on 9 and Qwen2 VL 72B Instruct on 2. The widest gap is mathvision, where Qwen2.5 VL 32B Instruct scores 40 against 25.9.

AlibabavsAlibaba11 shared benchmarks9–2 head-to-head
BenchmarkQwen2.5 VL 32B InstructQwen2 VL 72B Instruct
CC-OCR77.168.7
docvqa94.896.5
InfoVQA83.484.5
mathvision4025.9
MathVista74.770.5
MMBench-Video1.91.7
MMMU7064.5
MMMU-Pro49.546.2
MMStar69.568.3
ocrbench_v259.146.1
Video-MME77.977.8

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.