Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Search on VisBrowse (test)
Loading...
34.7
Accuracy
Visual-Seeker
7.244
14.372
21.5
28.628
Jun 13, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Visual-Seeker
Workflow Protocol=Mult...
2026.06
34.7
GPT-5
Workflow Protocol=Agen...
2026.06
29
Gemini-2.5-Pro
Workflow Protocol=Agen...
2026.06
26.6
GPT-5
Workflow Protocol=Dire...
2026.06
26
Gemini-2.5-Flash
Workflow Protocol=Agen...
2026.06
20.7
Gemini-2.5-Pro
Workflow Protocol=Dire...
2026.06
19.5
Claude-4-Sonnet
Workflow Protocol=Agen...
2026.06
19.5
Gemini-2.5-Flash
Workflow Protocol=Dire...
2026.06
16
Qwen3-VL-8B-Instruct
Workflow Protocol=Agen...
2026.06
15.4
Qwen3-VL-8B-Instruct
Workflow Protocol=Dire...
2026.06
10.7
Claude-4-Sonnet
Workflow Protocol=Dire...
2026.06
8.3
Feedback
Search any
task
Search any
task