Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Secure Code Generation on CYBERSECEVAL
Loading...
79.38
Autocomplete
Qwen-3-8B
67.4408
70.5404
73.64
76.7396
Oct 15, 2025
Autocomplete
Instruct
Average
Updated 23d ago
Evaluation Results
Method
Method
Links
Autocomplete
Instruct
Average
Qwen-3-8B
2025.10
79.38
79.77
79.58
Qwen-3-4B
2025.10
78.81
80.13
79.47
Qwen-3-14B
2025.10
77.45
74.54
75.99
CyberPal 2.0
Backbone=Qwen-3-4B
2025.10
76.1
67.16
71.63
CyberPal 2.0
Backbone=Qwen-3-14B
2025.10
73.33
66.33
69.83
CyberPal 2.0
Backbone=Qwen-3-8B
2025.10
72.81
69.07
70.94
gpt-oss-20b
2025.10
70.25
68.53
69.39
CyberPal 2.0
Backbone=gpt-oss-20b
2025.10
67.9
63.95
65.93
Feedback
Search any
task
Search any
task