Agentic Search on MMSearch v1.0 (test)
70.8AccuracyPOINTS-Seeker-8B
Evaluation Results
| Method | Links | |
|---|---|---|
| POINTS-Seeker-8BWorkflow=Agentic Search MLLM2026.04 | 70.8 | |
| Vision-DeepResearch-8BWorkflow=Agentic Search MLLM2026.04 | 69.6 | |
| SenseNova-MARS-8BWorkflow=Agentic Search MLLM2026.04 | 67.8 | |
| MM-DeepResearch-8BWorkflow=Agentic Search MLLM2026.04 | 67.8 | |
| Claude-4-SonnetWorkflow=Agent Workflow2026.04 | 67.2 | |
| Skywork-R1V4-30B-A3BWorkflow=Agentic Search MLLM2026.04 | 66.1 | |
| Gemini-2.5 FlashWorkflow=Agent Workflow2026.04 | 64 | |
| GPT-5Workflow=Agent Workflow2026.04 | 63.7 | |
| DeepEyesV2Workflow=Agentic Search MLLM2026.04 | 63.7 | |
| WebWatcher-32BWorkflow=Agentic Search MLLM2026.04 | 55.3 | |
| MMSearch-R1-7BWorkflow=Agentic Search MLLM2026.04 | 53.8 | |
| WebWatcher-7BWorkflow=Agentic Search MLLM2026.04 | 49.1 | |
| Qwen3-VL-8B-ThinkingWorkflow=Agent Workflow2026.04 | 45.6 | |
| Gemini-2.5 FlashWorkflow=RAG Workflow2026.04 | 43.9 | |
| GPT-5Workflow=Direct Answer2026.04 | 33.3 | |
| Claude-3.7-SonnetWorkflow=RAG Workflow2026.04 | 32.7 | |
| Gemini-2.5 FlashWorkflow=Direct Answer2026.04 | 30.4 | |
| Qwen-2.5-VL-72BWorkflow=RAG Workflow2026.04 | 29.2 | |
| GPT-4oWorkflow=RAG Workflow2026.04 | 24.1 | |
| Claude-4-SonnetWorkflow=Direct Answer2026.04 | 18.7 | |
| Qwen3-VL-8B-InstructWorkflow=Direct Answer2026.04 | 15.2 |