Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Safety Evaluation on MultiJail-AR (test)
Loading...
89.101
Safety Rate (%)
AWQ-trust
22.0782
39.47835
56.8785
74.27865
Jan 17, 2026
Safety Rate (%)
Updated 4d ago
Evaluation Results
Method
Method
Links
Safety Rate (%)
AWQ-trust
Backbone=Llama-3.1-8B-...
2026.01
89.101
AWQ-trust
Backbone=Qwen-2.5-7B-I...
2026.01
89.101
Full Precision
Backbone=Qwen-2.5-7B-I...
2026.01
81.376
AWQ
Backbone=Qwen-2.5-7B-I...
2026.01
77.566
AWQ-trust
Backbone=Gemma-7B-Inst...
2026.01
62.434
Full Precision
Backbone=Gemma-7B-Inst...
2026.01
57.143
AWQ
Backbone=Gemma-7B-Inst...
2026.01
55.979
Full Precision
Backbone=Llama-3.1-8B-...
2026.01
31.852
AWQ
Backbone=Llama-3.1-8B-...
2026.01
24.656
Feedback
Search any
task
Search any
task