List your company

Web-search QA on VDR

33.8Pass@1OpenSearch-VL-32B
0.4169.08317.7526.417May 6, 2026

Evaluation results

MethodLinks
OpenSearch-VL-32BWorkflow=Agentic Workflow2026.0533.8
OpenSearch-VL-30B-A3BWorkflow=Agentic Workflow2026.0533.5
Qwen3-VL-32BWorkflow=Agentic Workflow2026.0523.1
GPT-5Workflow=RAG Workflow2026.0522.3
OpenSearch-VL-8BWorkflow=Agentic Workflow2026.0520.8
Qwen3-VL-30B-A3BWorkflow=Agentic Workflow2026.0520.2
SenseNova-MARS-8BWorkflow=Agentic Workflow2026.0519.4
Qwen3-VL-8BWorkflow=Agentic Workflow2026.0517
Claude-3.7-SonnetWorkflow=RAG Workflow2026.0511.3
WebWatcher-7BWorkflow=Agentic Workflow2026.0510.3
GPT-5Workflow=Direct Reasoning2026.059.8
Gemini-2.5-ProWorkflow=Direct Reasoning2026.058
DeepEyes-v2-7BWorkflow=Agentic Workflow2026.057.8
Qwen3-VL-8BWorkflow=RAG Workflow2026.057.3
Gemini-2.5-FlashWorkflow=Direct Reasoning2026.056.2
Claude-3.7-SonnetWorkflow=Direct Reasoning2026.054.6
GPT-4oWorkflow=RAG Workflow2026.054.5
Qwen3-VL-32BWorkflow=Direct Reasoning2026.054.1
Qwen3-VL-30B-A3BWorkflow=Direct Reasoning2026.053.8
Visual-ARFT-7BWorkflow=Agentic Workflow2026.053.3
MMSearch-R1-7BWorkflow=Agentic Workflow2026.052.9
Qwen3-VL-8BWorkflow=Direct Reasoning2026.052.8
Claude-4-SonnetWorkflow=Direct Reasoning2026.052
GPT-4oWorkflow=Direct Reasoning2026.051.7