List your company

Web-search QA on BrowseComp-VL

54.9Pass@1GPT-5
3.52416.86230.243.538May 6, 2026

Evaluation results

MethodLinks
GPT-5Workflow=RAG Workflow2026.0554.9
GPT-5Workflow=Direct Reasoning2026.0548.6
OpenSearch-VL-32BWorkflow=Agentic Workflow2026.0543.8
Gemini-2.5-ProWorkflow=Direct Reasoning2026.0543.1
OpenSearch-VL-30B-A3BWorkflow=Agentic Workflow2026.0541.1
OpenSearch-VL-8BWorkflow=Agentic Workflow2026.0537.6
Gemini-2.5-FlashWorkflow=Direct Reasoning2026.0537.1
SenseNova-MARS-8BWorkflow=Agentic Workflow2026.0535.1
Qwen3-VL-32BWorkflow=Agentic Workflow2026.0535.1
Qwen3-VL-30B-A3BWorkflow=Agentic Workflow2026.0534.1
Claude-3.7-SonnetWorkflow=Direct Reasoning2026.0532.3
Qwen3-VL-32BWorkflow=Direct Reasoning2026.0530.8
Qwen3-VL-30B-A3BWorkflow=Direct Reasoning2026.0529.6
Claude-4-SonnetWorkflow=Direct Reasoning2026.0529.3
Qwen3-VL-8BWorkflow=RAG Workflow2026.0529.3
Qwen3-VL-8BWorkflow=Agentic Workflow2026.0527.9
WebWatcher-32BWorkflow=Agentic Workflow2026.0527
Qwen3-VL-8BWorkflow=Direct Reasoning2026.0524.1
WebWatcher-7BWorkflow=Agentic Workflow2026.0521.2
MMSearch-R1-7BWorkflow=Agentic Workflow2026.0520.9
Visual-ARFT-7BWorkflow=Agentic Workflow2026.0516.5
GPT-4oWorkflow=RAG Workflow2026.0513.4
Claude-3.7-SonnetWorkflow=RAG Workflow2026.0510
GPT-4oWorkflow=Direct Reasoning2026.055.5