Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Safety Detection on SafeDialBench (full)
Loading...
99
Recall
Nemotron
13.928
36.014
58.1
80.186
Jan 1, 2026
Recall
Tokens
Updated 4d ago
Evaluation Results
Method
Method
Links
Recall
Tokens
Nemotron
Training=Baseline
2026.01
99
3,020
Qwen3Guard
Training=M2S Hyphenize
2026.01
93.8
173
Nemotron
Training=M2S Numberize
2026.01
87.8
177
Nemotron
Training=M2S Pythonize
2026.01
82.9
288
LlamaGuard
Training=Baseline
2026.01
75.1
3,110
Nemotron
Training=M2S Hyphenize
2026.01
67.6
176
Qwen3Guard
Training=Baseline
2026.01
54.9
3,231
Qwen3Guard
Training=M2S Numberize
2026.01
33.6
174
Qwen3Guard
Training=M2S Pythonize
2026.01
30.6
285
LlamaGuard
Training=M2S Numberize
2026.01
24.5
176
LlamaGuard
Training=M2S Hyphenize
2026.01
24.1
175
LlamaGuard
Training=M2S Pythonize
2026.01
17.2
287
Feedback
Search any
task
Search any
task