Qwen 2.5 VL 72B Instruct — benchmark results
Alibaba Qwen 2.5 VL 72B vision-language instruction-tuned checkpoint. Provider: Alibaba. Released 2025-01-26. Access: Open.
Unified ELO 1575 ± 16, rank #604 of 1839 rated models, from 75 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Stats (DocVQA) | 96.4 | Score (%) | 100 |
| StickToYourRole | 83.6 | Cardinal Score | 100 |
| TableVista | 55.8 | Avg. (self-reported) | 96.4 |
| Open Portuguese LLM - Hate Speech | 76.96 | F1 (%) | 93.9 |
| KOFFVQA - Korean OCR | 95 | Score (%) | 93.8 |
| LLM Stats (ChartQA) | 89.5 | Score (%) | 91.3 |
| CapArena-Auto | 35.3 | CapArena-Auto Score (avg) | 90.5 |
| CapArena-Auto vs CogVLM-19B | 49 | CapArena-Auto Score (vs CogVLM-19B) | 90.5 |
| CapArena-Auto vs GPT-4o | -1 | CapArena-Auto Score (vs GPT-4o) | 90.5 |
| LLM Stats (EgoSchema) | 76.2 | Score (%) | 87.5 |
| LLM Stats (MMBench) | 88 | Score (%) | 87.5 |
| KOFFVQA - Recognition | 90 | Score (%) | 87 |
Interactive version: theaggregate.ai/model?slug=qwen-2-5-vl-72b-instruct · How It Works · Data refreshed daily, snapshot 2026-08-05.