Share your thoughts, 1 month free Claude Pro on usSee more

Agentic Search on BC-VL

48.6Accuracy

Claude-4-Sonnet

Updated 2d ago

Evaluation Results

Method	Links
Claude-4-Sonnet 2026.04		48.6
GPT-5 2026.04		47.2
GPT-5 2026.04		46.1
Gemini-2.5 Flash 2026.04		44.6
POINTS-Seeker-8B 2026.04		44.4
Vision-DeepResearch-8B 2026.04		42.6
Skywork-R1V4-30B-A3B 2026.04		38.4
MM-DeepResearch-8B 2026.04		37.9
Gemini-2.5 Flash 2026.04		37.1
Qwen3-VL-8B-Thinking 2026.04		37.1
Claude-4-Sonnet 2026.04		29.3
WebWatcher-32B 2026.04		27
Qwen3-VL-8B-Instruct 2026.04		25.1
WebWatcher-7B 2026.04		21.2
GPT-4o 2026.04		13.4
Gemini-2.5 Flash 2026.04		13
Qwen-2.5-VL-72B 2026.04		11.5
Claude-3.7-Sonnet 2026.04		11.2