Multimodal Search on MMSearch (Pass@1)
72.3Pass@1OpenSearch-VL-32B
Evaluation Results
| Method | Links | |
|---|---|---|
| OpenSearch-VL-32BWorkflow=Agentic Workflow2026.05 | 72.3 | |
| OpenSearch-VL-30B-A3BWorkflow=Agentic Workflow2026.05 | 68.7 | |
| SenseNova-MARS-8BWorkflow=Agentic Workflow2026.05 | 67.4 | |
| OpenSearch-VL-8BWorkflow=Agentic Workflow2026.05 | 64.5 | |
| DeepEyes-v2-7BWorkflow=Agentic Workflow2026.05 | 63.7 | |
| WebWatcher-32BWorkflow=Agentic Workflow2026.05 | 55.3 | |
| Qwen3-VL-32BWorkflow=Agentic Workflow2026.05 | 53.9 | |
| MMSearch-R1-7BWorkflow=Agentic Workflow2026.05 | 53.8 | |
| GPT-5Workflow=RAG Workflow2026.05 | 52.6 | |
| GPT-4oWorkflow=RAG Workflow2026.05 | 49.1 | |
| WebWatcher-7BWorkflow=Agentic Workflow2026.05 | 49.1 | |
| Qwen3-VL-8BWorkflow=RAG Workflow2026.05 | 47.3 | |
| Qwen3-VL-30B-A3BWorkflow=Agentic Workflow2026.05 | 44.2 | |
| Gemini-2.5-ProWorkflow=Direct Reasoning2026.05 | 39.8 | |
| Qwen3-VL-8BWorkflow=Agentic Workflow2026.05 | 37.4 | |
| GPT-5Workflow=Direct Reasoning2026.05 | 35.1 | |
| Visual-ARFT-7BWorkflow=Agentic Workflow2026.05 | 34.5 | |
| Claude-3.7-SonnetWorkflow=RAG Workflow2026.05 | 32.7 | |
| Gemini-2.5-FlashWorkflow=Direct Reasoning2026.05 | 30.4 | |
| Claude-3.7-SonnetWorkflow=Direct Reasoning2026.05 | 21.1 | |
| Qwen3-VL-32BWorkflow=Direct Reasoning2026.05 | 19.8 | |
| GPT-4oWorkflow=Direct Reasoning2026.05 | 18.7 | |
| Claude-4-SonnetWorkflow=Direct Reasoning2026.05 | 18.7 | |
| Qwen3-VL-30B-A3BWorkflow=Direct Reasoning2026.05 | 18.7 | |
| Qwen3-VL-8BWorkflow=Direct Reasoning2026.05 | 11.7 |