Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Comprehensive Evaluation on VizWiz (val)
Loading...
69.16
Score
MetaGPT
22.568
34.664
46.76
58.856
Mar 31, 2025
Score
Updated 3d ago
Evaluation Results
Method
Method
Links
Score
MetaGPT
merging_target=Qwen2-V...
2025.03
69.16
AdaMMS
merging_target=Qwen2-V...
2025.03
68.4
Qwen2-VL (base)
model_status=Original...
2025.03
68.32
Ties-Merging
merging_target=Qwen2-V...
2025.03
66.34
Task Arithmetic
merging_target=Qwen2-V...
2025.03
62.76
LLaVA-OneVision
model_status=Original...
2025.03
60.97
DARE-Ties
merging_target=Qwen2-V...
2025.03
58.05
LLaVA-v1.5-7B
Status=Source Model
2025.03
54.29
DARE-Linear
merging_target=Qwen2-V...
2025.03
50.18
AdaMMS
Merging Protocol=AdaMMS
2025.03
37.27
CogVLM-chat-7B
Status=Base Model
2025.03
37.09
MetaGPT
Merging Protocol=MetaGPT
2025.03
36.97
DARE-Linear
Merging Protocol=DARE-...
2025.03
36.12
Task Arithmetic
Merging Protocol=Task...
2025.03
35.7
DARE-Ties
Merging Protocol=DARE-...
2025.03
31.43
Ties-Merging
Merging Protocol=Ties-...
2025.03
24.36
Feedback
Search any
task
Search any
task