Web-search QA on VDR
33.8Pass@1OpenSearch-VL-32B
Evaluation Results
| Method | Links | |
|---|---|---|
| OpenSearch-VL-32BWorkflow=Agentic Workflow2026.05 | 33.8 | |
| OpenSearch-VL-30B-A3BWorkflow=Agentic Workflow2026.05 | 33.5 | |
| Qwen3-VL-32BWorkflow=Agentic Workflow2026.05 | 23.1 | |
| GPT-5Workflow=RAG Workflow2026.05 | 22.3 | |
| OpenSearch-VL-8BWorkflow=Agentic Workflow2026.05 | 20.8 | |
| Qwen3-VL-30B-A3BWorkflow=Agentic Workflow2026.05 | 20.2 | |
| SenseNova-MARS-8BWorkflow=Agentic Workflow2026.05 | 19.4 | |
| Qwen3-VL-8BWorkflow=Agentic Workflow2026.05 | 17 | |
| Claude-3.7-SonnetWorkflow=RAG Workflow2026.05 | 11.3 | |
| WebWatcher-7BWorkflow=Agentic Workflow2026.05 | 10.3 | |
| GPT-5Workflow=Direct Reasoning2026.05 | 9.8 | |
| Gemini-2.5-ProWorkflow=Direct Reasoning2026.05 | 8 | |
| DeepEyes-v2-7BWorkflow=Agentic Workflow2026.05 | 7.8 | |
| Qwen3-VL-8BWorkflow=RAG Workflow2026.05 | 7.3 | |
| Gemini-2.5-FlashWorkflow=Direct Reasoning2026.05 | 6.2 | |
| Claude-3.7-SonnetWorkflow=Direct Reasoning2026.05 | 4.6 | |
| GPT-4oWorkflow=RAG Workflow2026.05 | 4.5 | |
| Qwen3-VL-32BWorkflow=Direct Reasoning2026.05 | 4.1 | |
| Qwen3-VL-30B-A3BWorkflow=Direct Reasoning2026.05 | 3.8 | |
| Visual-ARFT-7BWorkflow=Agentic Workflow2026.05 | 3.3 | |
| MMSearch-R1-7BWorkflow=Agentic Workflow2026.05 | 2.9 | |
| Qwen3-VL-8BWorkflow=Direct Reasoning2026.05 | 2.8 | |
| Claude-4-SonnetWorkflow=Direct Reasoning2026.05 | 2 | |
| GPT-4oWorkflow=Direct Reasoning2026.05 | 1.7 |