Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-choice Visual Question Answering on camroll
Loading...
86.4
Accuracy
Oracle
27.744
42.972
58.2
73.428
Jun 3, 2026
Accuracy
Recall
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Recall
Oracle
Base Model=Gemini-2.5-...
2026.06
86.4
100
All images
Base Model=Gemini-2.5-...
2026.06
76.5
100
camroll-agent
Base Model=Gemini-2.5-...
2026.06
70.5
88.5
All captions
Base Model=Gemini-2.5-...
2026.06
63.4
100
ClaudeCode
Base Model=Sonnet-4.6,...
2026.06
54
-
Mem0
Base Model=Gemini-2.5-...
2026.06
53.2
75.3
LightMem
Base Model=Gemini-2.5-...
2026.06
52.7
70.3
HippoRAG2
Base Model=Gemini-2.5-...
2026.06
48.5
50.1
Self-RAG
Base Model=LLama-2, Re...
2026.06
46.2
25.8
SimpleMem
Base Model=Gemini-2.5-...
2026.06
44.6
57.8
MemOS
Base Model=Gemini-2.5-...
2026.06
32.3
27.5
Nothing
Base Model=Gemini-2.5-...
2026.06
30
0
Feedback
Search any
task
Search any
task