Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Guardrail evaluation on Representative guardrail task ~1250 tokens
Loading...
0.01
Evaluation Cost per 1K
Luna-2
-0.3176
1.8937
4.105
6.3163
Feb 20, 2026
Evaluation Cost per 1K
Latency (ms)
Updated 4d ago
Evaluation Results
Method
Method
Links
Evaluation Cost per 1K
Latency (ms)
Luna-2
Model size=3B
2026.02
0.01
150
GPT 4.1 mini
2026.02
0.75
2,800
Azure Content Safety
2026.02
1.65
312
GPT 4.1
2026.02
3.6
3,000
Gemini 3 Pro
Thinking mode=low
2026.02
5.85
6,900
Claude Sonnet 4.5
2026.02
8.2
6,700
Feedback
Search any
task
Search any
task