Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Video Spatial Understanding on MMSI-Bench 500 QA 1.0
Loading...
41
Accuracy
CoCoSI
30.808
33.454
36.1
38.746
Jun 9, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
CoCoSI
Backbone=Gemini-based,...
2026.06
41
CoCoSI
Backbone=GPT-based, Tr...
2026.06
39.1
Single-Agent Baseline
Backbone=Gemini-based,...
2026.06
36.6
TraveLER
Backbone=Gemini-based,...
2026.06
36
VCA
Backbone=Gemini-based,...
2026.06
35.8
VideoAgent
Backbone=Gemini-based,...
2026.06
35.4
CoCoSI
Backbone=Qwen-VL-based...
2026.06
35.3
Single-Agent Baseline
Backbone=GPT-based, Tr...
2026.06
35
TraveLER
Backbone=GPT-based, Tr...
2026.06
34.7
VCA
Backbone=GPT-based, Tr...
2026.06
34.2
VideoAgent
Backbone=GPT-based, Tr...
2026.06
33.8
Single-Agent Baseline
Backbone=Qwen-VL-based...
2026.06
32.4
TraveLER
Backbone=Qwen-VL-based...
2026.06
31.9
VCA
Backbone=Qwen-VL-based...
2026.06
31.6
VideoAgent
Backbone=Qwen-VL-based...
2026.06
31.2
Feedback
Search any
task
Search any
task