Tool Use on VerlTool OOD Tools
404AttributeDeepEyesV2-7B
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| DeepEyesV2-7B2026.06 | 404 | 114 | 260 | 30 | 28.2 | 68.1 | 2.6 | 2.44 | 642.37 | |
| TikArt-8B2026.06 | 292 | 161 | 100 | 31 | 55.1 | 80.27 | 2.56 | 7.7 | 383.35 | |
| Qwen2.5-VL-7B2026.06 | 230 | 131 | 83 | 16 | 57 | 71.42 | 1.07 | 0.97 | 174.28 | |
| AdaReasoner-7B2026.06 | 200 | 50 | 132 | 18 | 25 | 72.9 | 2.52 | 4.39 | 312.68 | |
| VTool-R1-7B2026.06 | 86 | 24 | 57 | 5 | 27.9 | 75.15 | 1.41 | 2.27 | 262.26 | |
| PixelReasoner-RL-v1-7B2026.06 | 67 | 33 | 22 | 12 | 49.3 | 76.83 | 1.58 | 1.82 | 306.81 | |
| ZwZ-8B2026.06 | 44 | 28 | 14 | 2 | 63.6 | 81.35 | 1.45 | 0.9 | 236.67 | |
| PyVision-Image-7B-RL2026.06 | 32 | 15 | 17 | 0 | 46.9 | 54.04 | 5.83 | 2.5 | 1,087.73 | |
| Qwen3-VL-8B2026.06 | 28 | 5 | 22 | 1 | 17.9 | 74.21 | 1.4 | 2.17 | 272.68 | |
| MetaForge2026.06 | 12 | 11 | 0 | 1 | 91.7 | 88.98 | 2.3 | 4.97 | 336.65 | |
| R1-Onevision-7B2026.06 | 6 | 3 | 3 | 0 | 50 | 65.76 | 1.01 | 0.01 | 304.99 |