Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Compound Editing on Real-world compound editing evaluation set (test)
Loading...
0.1868
L1
Qwen
0.18502
0.197035
0.20905
0.221065
Apr 6, 2026
L1
L2
CLIP
DINO
Updated 12d ago
Evaluation Results
Method
Method
Links
L1
L2
CLIP
DINO
Qwen
2026.04
0.1868
0.0603
0.864
0.7896
HorizonWeaver
2026.04
0.1955
0.0708
0.874
0.8232
UltraEdit
2026.04
0.2208
0.0848
0.8667
0.7838
Bagel
2026.04
0.2258
0.0899
0.8224
0.6918
OmniGen2
2026.04
0.2313
0.0948
0.8385
0.7212
Feedback
Search any
task
Search any
task