Multimodal Question Answering on LiveVQA
70.5Pass@1OpenSearch-VL-32B
Evaluation Results
| Method | Links | |
|---|---|---|
| OpenSearch-VL-32BWorkflow=Agentic Workflow2026.05 | 70.5 | |
| OpenSearch-VL-30B-A3BWorkflow=Agentic Workflow2026.05 | 67.4 | |
| Qwen3-VL-30B-A3BWorkflow=Agentic Workflow2026.05 | 62 | |
| Gemini-2.5-ProWorkflow=Direct Reasoning2026.05 | 60.3 | |
| OpenSearch-VL-8BWorkflow=Agentic Workflow2026.05 | 59.6 | |
| WebWatcher-32BWorkflow=Agentic Workflow2026.05 | 58.7 | |
| SenseNova-MARS-8BWorkflow=Agentic Workflow2026.05 | 56.2 | |
| GPT-5Workflow=RAG Workflow2026.05 | 56 | |
| WebWatcher-7BWorkflow=Agentic Workflow2026.05 | 51.2 | |
| Gemini-2.5-FlashWorkflow=Direct Reasoning2026.05 | 51 | |
| Qwen3-VL-8BWorkflow=Agentic Workflow2026.05 | 50.6 | |
| MMSearch-R1-7BWorkflow=Agentic Workflow2026.05 | 48.4 | |
| Qwen3-VL-32BWorkflow=Direct Reasoning2026.05 | 45.5 | |
| Qwen3-VL-32BWorkflow=Agentic Workflow2026.05 | 45.5 | |
| GPT-5Workflow=Direct Reasoning2026.05 | 44.4 | |
| Qwen3-VL-30B-A3BWorkflow=Direct Reasoning2026.05 | 42.7 | |
| GPT-4oWorkflow=RAG Workflow2026.05 | 40.1 | |
| Qwen3-VL-8BWorkflow=RAG Workflow2026.05 | 39.3 | |
| Claude-4-SonnetWorkflow=Direct Reasoning2026.05 | 38.5 | |
| Claude-3.7-SonnetWorkflow=Direct Reasoning2026.05 | 38 | |
| Claude-3.7-SonnetWorkflow=RAG Workflow2026.05 | 30.3 | |
| GPT-4oWorkflow=Direct Reasoning2026.05 | 28.1 | |
| Visual-ARFT-7BWorkflow=Agentic Workflow2026.05 | 25.4 | |
| Qwen3-VL-8BWorkflow=Direct Reasoning2026.05 | 23.1 |