Perception on TreeBench
55.3Overall AccuracyPFlowNet
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| PFlowNetBase Model=Qwen3-VL-8B2026.05 | 55.3 | 65.5 | 69.2 | 80.2 | 75 | 77.9 | |
| TreeVGR2026.05 | 50.4 | 65.5 | 53.8 | 82.6 | 68.8 | 63.3 | |
| InternVL3-78BBackbone Parameters=78B2026.05 | 46.4 | 62.1 | 61.5 | 52.2 | 68.8 | 52.9 | |
| Qwen3-VL-32BBackbone Parameters=32B2026.05 | 45.2 | 60.3 | 63.4 | 58.1 | 83.6 | 30.3 | |
| Qwen3-VL-8BBackbone Parameters=8B2026.05 | 44.9 | 65.5 | 53.9 | 65.2 | 75 | 64.7 | |
| Qwen2.5-VL-32BBackbone Parameters=32B2026.05 | 42.5 | 51.7 | 53.8 | 69.6 | 62.5 | 54.4 | |
| Qwen2.5-VL-72BBackbone Parameters=72B2026.05 | 42.2 | 65.5 | 69.2 | 56.5 | 56.3 | 48.5 | |
| Qwen3-VL-4BBackbone Parameters=4B2026.05 | 42.2 | 48.3 | 61.5 | 65.2 | 81.3 | 35.3 | |
| InternVL3-38BBackbone Parameters=38B2026.05 | 42 | 51.7 | 61.5 | 52.2 | 68.8 | 51.5 | |
| DeepEyesV22026.05 | 40.7 | 65.5 | 69.2 | 56.5 | 62.5 | 55.9 | |
| LLaVA-OV-72BBackbone Parameters=72B2026.05 | 40.5 | 62.1 | 53.8 | 65.2 | 62.3 | 36.8 | |
| Pixel-Reasoner2026.05 | 39 | 58.6 | 61.5 | 65.2 | 50 | 48.5 | |
| InternVL3-8BBackbone Parameters=8B2026.05 | 38.8 | 51.7 | 69.2 | 56.5 | 56.3 | 33.7 | |
| Thyme2026.05 | 38.2 | 48.2 | 46.1 | 69.5 | 50 | 51.4 | |
| DeepEyes2026.05 | 37.5 | 62.1 | 53.8 | 65.2 | 68.8 | 51.5 | |
| LLaVA-OV-7BBackbone Parameters=7B2026.05 | 37.3 | 55.2 | 53.8 | 56.5 | 50 | 32.4 | |
| Qwen2.5-VL-7BBackbone Parameters=7B2026.05 | 37 | 55.2 | 53.8 | 56.5 | 62.5 | 27.9 |