Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Human Preference Alignment on SD3.5 Medium
Loading...
0.361
HPS-v2.1
TMPO
0.26844
0.29247
0.3165
0.34053
May 9, 2026
HPS-v2.1
ImgRwd
PickScore
LGMD
Cosine Divergence
Updated 21d ago
Evaluation Results
Method
Method
Links
HPS-v2.1
ImgRwd
PickScore
LGMD
Cosine Divergence
TMPO
Time (s)=63.9, Fine-tu...
2026.05
0.361
1.358
23.551
0.118
0.239
Flow-GRPO
Time (s)=101.8, Fine-t...
2026.05
0.355
1.329
23.289
-0.123
0.195
TreeGRPO
Time (s)=74.6, Fine-tu...
2026.05
0.351
1.298
22.817
-0.289
0.168
MixGRPO
Time (s)=96.2, Fine-tu...
2026.05
0.348
1.351
23.019
-0.238
0.175
GARDO
Time (s)=105.1, Fine-t...
2026.05
0.329
1.335
23.102
0.072
0.227
SD3.5-Medium
Training=Baseline
2026.05
0.272
0.893
21.592
-0.061
0.213
Feedback
Search any
task
Search any
task