Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Preference Alignment on AlpacaEval weighted gpt4 turbo 2.0
Loading...
46.11
Win Rate
GANPO (SimPO)
21.826
28.1305
34.435
40.7395
Jan 29, 2026
Win Rate
LC Win Rate
Avg Response Length
Updated 4d ago
Evaluation Results
Method
Method
Links
Win Rate
LC Win Rate
Avg Response Length
GANPO (SimPO)
Backbone=Llama3-8B-Ins...
2026.01
46.11
50.48
1,834
SimPO
Backbone=Llama3-8B-Ins...
2026.01
44.09
48.31
1,836
GANPO (DPO)
Backbone=Llama3-8B-Ins...
2026.01
35.23
33.87
2,043
DPO
Backbone=Llama3-8B-Ins...
2026.01
33.9
32.34
2,041
GANPO (SimPO)
Backbone=Gemma2-2B-it,...
2026.01
31.37
36.74
1,745
SimPO
Backbone=Gemma2-2B-it
2026.01
30.66
36.03
1,740
GANPO (DPO)
Backbone=Gemma2-2B-it,...
2026.01
24.17
29.69
1,664
DPO
Backbone=Gemma2-2B-it
2026.01
22.76
27.79
1,668
Feedback
Search any
task
Search any
task