WebQA on WebMMU
91.06Agentic Action Success RateQwen 2.5 VL 72B
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen 2.5 VL 72BModel Category=Open-Source Models, Backbone=72B2026.01 | 91.06 | |
| GPT 5 + ToolsModel Category=Tool-Planning Models, Tool Usage=true2026.01 | 88.82 | |
| Qwen 2.5 VL 32BModel Category=Open-Source Models, Backbone=32B2026.01 | 85.98 | |
| Claude 4 sonnetModel Category=Closed-Source Models2026.01 | 83.54 | |
| GPT 5Model Category=Closed-Source Models2026.01 | 80.49 | |
| Qwen 2.5 VL 72B + ToolsModel Category=Tool-Planning Models, Tool Usage=true, Backbone=72B2026.01 | 76.83 | |
| DeepEyesModel Category=Tool-Planning Models, Unified Evaluation Framework=true2026.01 | 72.76 | |
| AdaReasoner 7BModel Category=Tool-Planning Models, Backbone=7B2026.01 | 72.15 | |
| InternVL3 78BModel Category=Open-Source Models, Backbone=78B2026.01 | 71.34 | |
| Qwen 2.5 VL 7B + ToolsModel Category=Tool-Planning Models, Tool Usage=true, Backbone=7B2026.01 | 69.51 | |
| PixelReasonerModel Category=Tool-Planning Models, Unified Evaluation Framework=true2026.01 | 69.51 | |
| Qwen 2.5 VL 7BModel Category=Open-Source Models, Backbone=7B2026.01 | 67.48 | |
| Gemini 2.5 flashModel Category=Closed-Source Models2026.01 | 66.26 | |
| Qwen 2.5 VL 3BModel Category=Open-Source Models, Backbone=3B2026.01 | 54.47 |