Multi-modal Search on MMSearch (Score)
74.27ScoreGemini-3-Pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini-3-ProWorkflow=Agentic Model, Zero-shot=true2026.07 | 74.27 | |
| VideoSearcher-8BWorkflow=Agentic Model, Zero-shot=false2026.07 | 67.84 | |
| GPT-5.2Workflow=Agentic Model, Zero-shot=true2026.07 | 66.08 | |
| VideoSearcher-4BWorkflow=Agentic Model, Zero-shot=false2026.07 | 66.08 | |
| DeepEyesV2Workflow=Agentic Model, Zero-shot=false2026.07 | 63.7 | |
| Gemini-3-ProWorkflow=Direct Answer, Zero-shot=false2026.07 | 62.57 | |
| Gemini-3-FlashWorkflow=Agentic Model, Zero-shot=true2026.07 | 62.57 | |
| Gemini-3-FlashWorkflow=Direct Answer, Zero-shot=false2026.07 | 57.31 | |
| MMSearch-R1Workflow=Agentic Model, Zero-shot=false2026.07 | 53.8 | |
| Qwen2.5-VL-32B-InstructWorkflow=Agentic Model, Zero-shot=true2026.07 | 49.71 | |
| Qwen3-VL-4B-InstructWorkflow=Agentic Model, Zero-shot=true2026.07 | 49.12 | |
| Qwen3-VL-32B-InstructWorkflow=Agentic Model, Zero-shot=true2026.07 | 49.12 | |
| GPT-4oWorkflow=Agentic Model, Zero-shot=true2026.07 | 49.12 | |
| Webwatcher-7BWorkflow=Agentic Model, Zero-shot=false2026.07 | 49.1 | |
| Qwen3-VL-8B-InstructWorkflow=Agentic Model, Zero-shot=true2026.07 | 47.37 | |
| GPT-5.2Workflow=Direct Answer, Zero-shot=false2026.07 | 43.27 | |
| Visual-ARFTWorkflow=Agentic Model, Zero-shot=false2026.07 | 34.5 | |
| Qwen2.5-VL-7B-InstructWorkflow=Agentic Model, Zero-shot=true2026.07 | 32.16 | |
| GPT-4oWorkflow=Direct Answer, Zero-shot=false2026.07 | 23.39 | |
| Qwen3-VL-32B-InstructWorkflow=Direct Answer, Zero-shot=false2026.07 | 16.96 | |
| Qwen3-VL-4B-InstructWorkflow=Direct Answer, Zero-shot=false2026.07 | 13.45 | |
| Qwen3-VL-8B-InstructWorkflow=Direct Answer, Zero-shot=false2026.07 | 11.7 | |
| Qwen2.5-VL-32B-InstructWorkflow=Direct Answer, Zero-shot=false2026.07 | 11.7 | |
| Qwen2.5-VL-7B-InstructWorkflow=Direct Answer, Zero-shot=false2026.07 | 7.6 |