DeepShop deepshop Leaderboard
Auto-discovered from papers reporting DeepShop (Success rate). · Metric: Success rate (higher is better) · 🟢 Updated 2h ago
| # | Model | Success rate | Paper |
|---|---|---|---|
| 1 | OpenWebRL: Demystifying Online Multi-turn Reinforcement Learning for Visual Web Agents | 64.00 | link |
| 2 | Orchard-GUI-4B (SFT + RL) | 64.00 | link |
| 3 | Gemini computer-use-preview | 62.00 | link |
| 4 | Qwen3-VL-235B-A22B-Thinking | 56.70 | link |
| 5 | o3 (SoM) | 49.70 | link |
| 6 | Orchard-GUI-4B-SFT | 48.70 | link |
| 7 | Gemini-3-flash (Axtree) [30 steps] | 45.10 | link |
| 8 | MolmoWeb-8B | 42.30 | link |
| 9 | GPT-5 (Axtree) | 40.70 | link |
| 10 | MolmoWeb-4B | 35.60 | link |
| 11 | Qwen3-VL-4B-Thinking | 33.30 | link |
| 12 | GLM-4.1V-9B-Thinking | 32.00 | link |
| 13 | Fara-7B | 26.20 | link |
| 14 | OpenAI computer-use-preview | 24.70 | link |
| 15 | GPT-4o (SoM) | 16.00 | link |
| 16 | UI-TARS-1.5-7B | 11.60 | link |