Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Evaluation on UltraFeedback Aggregate
Loading...
59.59
Overall Average Score
LP-SFT
20.3508
30.5379
40.725
50.9121
Jul 6, 2026
Overall Average Score
Updated 18d ago
Evaluation Results
Method
Method
Links
Overall Average Score
LP-SFT
Backbone=Qwen3-14B
2026.07
59.59
ASFT
Backbone=Qwen3-14B
2026.07
57.43
EAFT
Backbone=Qwen3-14B
2026.07
55.94
CE
Backbone=Qwen3-14B
2026.07
55.54
LP-SFT
Backbone=Qwen3-4B
2026.07
54.74
GEM
Backbone=Qwen3-14B
2026.07
53.55
EAFT
Backbone=Qwen3-4B
2026.07
51
ASFT
Backbone=Qwen3-4B
2026.07
50.53
CE
Backbone=Qwen3-4B
2026.07
49.9
GEM
Backbone=Qwen3-4B
2026.07
49.39
DFT
Backbone=Qwen3-14B
2026.07
48.47
DFT
Backbone=Qwen3-4B
2026.07
44.02
LP-SFT
Backbone=Llama3.1-8B
2026.07
34.01
GEM
Backbone=Llama3.1-8B
2026.07
32.86
CE
Backbone=Llama3.1-8B
2026.07
32.74
EAFT
Backbone=Llama3.1-8B
2026.07
31.79
ASFT
Backbone=Llama3.1-8B
2026.07
30.71
DFT
Backbone=Llama3.1-8B
2026.07
21.86
Feedback
Search any
task
Search any
task