Qwen2.5 VL 72B Instruct

Qwen / Qwen

31scores
30benchmarks
$0.25 / $0.75 per 1M tokenscost in/out

Metadata

Qwen Open source

Aliases: qwen-qwen2.5-vl-72b-instruct, qwen/qwen2.5-vl-72b-instruct, qwen2.5-vl-72b-instruct

Benchmark Results

Benchmark Category Rank Score Sampled
AITZ_EM Agentic 1 0.83 2026-05-06
Android Control High_EM Agentic 2 0.67 2026-05-06
Android Control Low_EM Agentic 1 0.94 2026-05-06
MobileMiniWob++_SR Agentic 2 0.68 2026-05-06
OSWorld Agentic 101 5.0% 2026-05-27
OSWorld Agentic 102 4.43% 2026-05-27
ScreenSpot-Pro Agentic 19 53.30 2026-05-06
OpenUGI Alignment 491 36.98 2026-05-06
Stick To Your Role! Alignment 1 0.84 2026-05-06
BizFinBench Finance 10 67.53 2026-05-27
FinEval Finance 18 71 2026-05-27
OmniEarth-Bench Geospatial 9 10.98 2026-05-27
MathVision Intelligence 82 38.10 2026-05-06
LatamBoard Multilingual 4 87.06 2026-05-06
ALL Bench Multimodal Multimodal 3 21.58 2026-05-06
CC-OCR Multimodal 12 0.80 2026-05-06
ChartQA Multimodal 3 0.90 2026-05-06
LVBench Multimodal 17 0.47 2026-05-06
Math-VR Multimodal 23 13.7 2026-05-27
MMBench-Video Multimodal 1 0.02 2026-05-06
MMLongBench-Doc Multimodal 15 35.20 2026-05-06
MMMU-Pro Multimodal 40 46.20 2026-05-06
MMSI-Bench Multimodal 15 30.7% 2026-05-28
MMVet Multimodal 1 0.76 2026-05-06
PerceptionTest Multimodal 1 0.73 2026-05-06
Physical AI Bench Understanding Multimodal 8 60.80 2026-05-06
TempCompass Multimodal 1 0.75 2026-05-06
Video SimpleQA Multimodal 8 39.50 2026-05-06
VideoMME w/o sub. Multimodal 8 0.73 2026-05-06
OCRBench-V2 (en) OCR 11 0.61 2026-05-06
K-MetBench Weather 26 67.1% accuracy 2026-05-28