Document Formatting on DOCFORMBENCH
2.69Average CallsUFO-2
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| UFO-2Model=GLM-4.6v, Model Category=Multimodal Models2026.06 | 2.69 | 0.21 | 8.05 | 37,362 | |
| WINDOWS-USEModel=Gemini-3-Flash-Preview, Model Category=Multimodal Models2026.06 | 3.26 | 1.15 | 7.96 | 12,466 | |
| DOCX-SKILLModel=GPT-5, Model Category=Multimodal Models2026.06 | 4 | 4.49 | 0.43 | 5,369 | |
| MCP-MSOFFICEModel=GPT-5, Model Category=Multimodal Models2026.06 | 4.37 | 12.32 | 0.88 | 2,651 | |
| DOCX-SKILLModel=Gemini-3-Flash-Preview, Model Category=Multimodal Models2026.06 | 4.67 | 23.55 | 0.95 | 8,351 | |
| WINDOWS-USEModel=Qwen3-VL-235B-A22B-I, Model Category=Multimodal Models2026.06 | 6.1 | 0.13 | 9.49 | 13,278 | |
| DOCX-SKILLModel=GLM-4.6v, Model Category=Multimodal Models2026.06 | 10.31 | 25.57 | 4.16 | 11,184 | |
| DOCX-SKILLModel=Qwen3-VL-235B-A22B-I, Model Category=Multimodal Models2026.06 | 13.39 | 15.53 | 8.7 | 10,819 | |
| MCP-MSOFFICEModel=Qwen3-VL-235B-A22B-I, Model Category=Multimodal Models2026.06 | 13.49 | 12.11 | 1.93 | 6,263 | |
| WINDOWS-USEModel=GLM-4.6v, Model Category=Multimodal Models2026.06 | 14.69 | 0.03 | 7.24 | 15,056 | |
| MCP-MSOFFICEModel=Gemini-3-Flash-Preview, Model Category=Multimodal Models2026.06 | 19.61 | 22.38 | 1.92 | 9,413 | |
| MCP-MSOFFICEModel=GLM-4.6v, Model Category=Multimodal Models2026.06 | 23.14 | 16.38 | 3.16 | 5,536 | |
| DOCX-SKILLModel=Qwen3-235B-A22B-I, Model Category=Language Models2026.06 | 24.22 | 12.36 | 2.71 | 13,372 | |
| MCP-MSOFFICEModel=Qwen3-235B-A22B-I, Model Category=Language Models2026.06 | 24.74 | 14.44 | 1.99 | 5,897 | |
| MCP-MSOFFICEModel=GLM-4.6, Model Category=Language Models2026.06 | 28.85 | 18.03 | 2 | 6,284 | |
| DOCFORMFLOWModel=Qwen3-VL-235B-A22B-I, Model Category=Multimodal Models2026.06 | 29.28 | 70.94 | 2.03 | 3,246 | |
| MCP-MSOFFICEModel=Deepseek-V3.2, Model Category=Language Models2026.06 | 30.02 | 15.41 | 1.61 | 9,534 | |
| DOCFORMFLOWModel=GLM-4.6v, Model Category=Multimodal Models2026.06 | 31.22 | 65.78 | 2.14 | 3,765 | |
| DOCFORMFLOWModel=Gemini-3-Flash-Preview, Model Category=Multimodal Models2026.06 | 31.26 | 80.36 | 1.58 | 4,187 | |
| DOCFORMFLOWModel=GLM-4.6, Model Category=Language Models2026.06 | 32.39 | 75.72 | 1.24 | 2,659 | |
| DOCFORMFLOWModel=Qwen3-235B-A22B-I, Model Category=Language Models2026.06 | 33.41 | 71.18 | 2.69 | 2,829 | |
| DOCFORMFLOWModel=GPT-5, Model Category=Multimodal Models2026.06 | 33.62 | 72.53 | 3.44 | 2,998 | |
| DOCFORMFLOWModel=Deepseek-V3.2, Model Category=Language Models2026.06 | 34.77 | 72.87 | 1.99 | 2,802 | |
| DOCX-SKILLModel=GLM-4.6, Model Category=Language Models2026.06 | 39.17 | 18.19 | 3.07 | 17,199 | |
| UFO-2Model=Gemini-3-Flash-Preview, Model Category=Multimodal Models2026.06 | 42.1 | 8.42 | 9.05 | 23,799 | |
| DOCX-SKILLModel=Deepseek-V3.2, Model Category=Language Models2026.06 | 42.14 | 20.71 | 1.64 | 17,074 | |
| WINDOWS-USEModel=GPT-5, Model Category=Multimodal Models2026.06 | 45.02 | 13.79 | 9.49 | 14,966 | |
| UFO-2Model=GPT-5, Model Category=Multimodal Models2026.06 | 45.4 | 16.91 | 2.07 | 26,165 | |
| UFO-2Model=Qwen3-VL-235B-A22B-I, Model Category=Multimodal Models2026.06 | 46.27 | 8.2 | 9.24 | 31,816 |