Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multi-discipline Reasoning on MMMU
Loading...
55.44
Accuracy
Vanilla
33.8392
39.4471
45.055
50.6629
Dec 5, 2024
Feb 13, 2025
Apr 25, 2025
Jul 5, 2025
Sep 13, 2025
Nov 23, 2025
Feb 2, 2026
Accuracy
Updated 2d ago
Evaluation Results
Method
Method
Links
Accuracy
Vanilla
Backbone=InternVL3-8B,...
2026.01
55.44
CAPA
Backbone=InternVL3-8B,...
2026.01
53.67
CAPA
Backbone=Qwen2.5-VL-7B...
2026.01
49.33
Vanilla
Backbone=Qwen2.5-VL-7B...
2026.01
48.68
LLaVA-v1.5-7B + Visual Lazy Attention
Base Model=LLaVA-v1.5-...
2026.02
36.8
Vanilla
Base Model=LLaVA 1.5 1...
2024.12
36.4
VisionZip
Base Model=LLaVA 1.5 1...
2024.12
36.4
VisionZip ‡
Base Model=LLaVA 1.5 1...
2024.12
36.4
VisionZip
Base Model=LLaVA 1.5 1...
2024.12
36.4
LLaVA-NEXT-13B + Visual Lazy Attention
Base Model=LLaVA-NEXT-...
2026.02
36.1
VisionZip
Base Model=LLaVA 1.5 1...
2024.12
36.1
CAPA
Backbone=LLaVA-1.5-7B,...
2026.01
35.89
VisionZip ‡
Base Model=LLaVA 1.5 1...
2024.12
35.4
LLaVA-v1.5-7B
Base Model=LLaVA-v1.5-...
2026.02
35.3
VisionZip ‡
Base Model=LLaVA 1.5 1...
2024.12
35.3
Vanilla
Backbone=LLaVA-1.5-7B,...
2026.01
34.67
Feedback
Search any
task
Search any
task