Qwen2.5-VL-72B-Instruct-AWQ: benchmark results
Provider: Alibaba. Access: Open.
Aggregate Rating 1634 ± 50, rank #730 of 10916 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PlanBench-V - Classification | 1.41 | Mean score (out of 2) on classifying planning maps by type w | 78.1 |
| PlanBench-V - Element Recognition | 1.3 | Mean score (out of 2) on element recognition (map layout, te | 75 |
| PlanBench-V - Caption | 1.83 | Mean score (out of 2) on map captioning, perception level; a | 71.9 |
| MedCUA-Bench | 6.2 | Strict success rate (%) over all 432 screenshot-only clinica | 62.5 |
| MedCUA-Bench - Intent Goals | 6.9 | Strict success rate (%) on the 216 intent-goal tasks (the cl | 62.5 |
| MedCUA-Bench - Step Goals | 5.6 | Strict success rate (%) on the 216 step-goal tasks (the same | 62.5 |
| PlanBench-V - Scheme Evaluation | 1.15 | Mean score (out of 2) on judging the strengths and weaknesse | 62.5 |
| PlanBench-V | 1.29 | Overall mean score (out of 2) on a 300-item stratified subse | 56.2 |
| PlanBench-V - Decision Making | 1.09 | Mean score (out of 2) on choosing between planning alternati | 50 |
| PlanBench-V - Domain-Specific Reasoning | 1.26 | Mean score (out of 2) on domain-specific planning reasoning | 50 |
| PlanBench-V - Policy Association | 1.25 | Mean score (out of 2) on linking map evidence to planning po | 50 |
| PlanBench-V - Spatial Relationship Reasoning | 1.25 | Mean score (out of 2) on topological, ordinal and metric spa | 50 |
Interactive version: theaggregate.ai/model?slug=qwen2-5-vl-72b-instruct-awq · How It Works · Data refreshed daily, snapshot 2026-11-08.