Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Model Alignment Evaluation on Re:Form SFT Evaluation Set (val)
Loading...
94.3
Validation Rate
14B SFT
-0.132
24.384
48.9
73.416
Jul 22, 2025
Validation Rate
Verification Rate
Spec Superiority Rate
Updated 19d ago
Evaluation Results
Method
Method
Links
Validation Rate
Verification Rate
Spec Superiority Rate
14B SFT
Model Size=14B, Traini...
2025.07
94.3
62.9
34.2
7B SFT
Model Size=7B, Trainin...
2025.07
90.8
53.3
27.9
3B SFT
Model Size=3B, Trainin...
2025.07
88.7
48
26.6
1.5B SFT
Model Size=1.5B, Train...
2025.07
84.2
41.6
22.1
0.5B SFT
Model Size=0.5B, Train...
2025.07
80.1
33.6
18
GPT-4o
2025.07
47.7
12.1
7
Qwen-Coder-7B
Model Size=7B, Trainin...
2025.07
17.6
3.7
0
Qwen-Coder-3B
Model Size=3B, Trainin...
2025.07
6.6
2.3
0.2
Qwen-Coder-14B
Model Size=14B, Traini...
2025.07
5.9
2.5
0.4
Qwen-Coder-1.5B
Model Size=1.5B, Train...
2025.07
5.5
1.2
0
Qwen-Coder-0.5B
Model Size=0.5B, Train...
2025.07
3.5
1.6
0
Feedback
Search any
task
Search any
task