Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Instruction-based Image Editing on GIER 50 user study samples
Loading...
0.225
L Metric
JarvisArt
-0.009
0.05175
0.1125
0.17325
Jun 6, 2026
L Metric
RL Metric
Updated 1mo ago
Evaluation Results
Method
Method
Links
L Metric
RL Metric
JarvisArt
Category=Tool-Calling...
2026.06
0.225
-0.328
Gemini-2.5-Pro
Category=Tool-Calling...
2026.06
0.225
-0.42
GPT-Image-1
Category=Generative Me...
2026.06
0.218
-0.34
GPT-4.1
Category=Tool-Calling...
2026.06
0.209
-0.306
Qwen2.5-VL-7B
Category=Tool-Calling...
2026.06
0.2
-0.221
Qwen-Image-Edit
Category=Generative Me...
2026.06
0.184
-0.132
IEA
Category=Tool-Calling...
2026.06
0.128
0.138
Reference
Type=Human Expert
2026.06
0
1
Feedback
Search any
task
Search any
task