Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multi-Image Fine-Grained Question Answering on MIFG-QA
Loading...
50.24
Accuracy (Nature)
IMAgent
35.7944
39.5447
43.295
47.0453
Dec 5, 2025
Accuracy (Nature)
Accuracy (Poster)
Average Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy (Nature)
Accuracy (Poster)
Average Accuracy
IMAgent
E2E=true, Param=7B
2025.12
50.24
48.73
49.21
GPT-4o
2025.12
49.26
47.55
48.11
Pixel Reasoner
E2E=false, Param=7B
2025.12
41.87
45.72
44.49
Deepeyes
E2E=true, Param=7B
2025.12
38.75
43.84
38.36
LLaVA-OneVision
Param=7B
2025.12
37.93
39.49
38.99
Qwen2.5-VL
Param=7B
2025.12
36.94
44.34
41.98
Qwen2.5-VL-CoT
Param=7B
2025.12
36.35
40.65
39.31
Feedback
Search any
task
Search any
task