Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Prompt-side safety risk classification on S-Eval
Loading...
94.3
F1 Score
Stage3-DPO
82.964
85.907
88.85
91.793
Jul 7, 2026
F1 Score
Updated 17d ago
Evaluation Results
Method
Method
Links
F1 Score
Stage3-DPO
Params=4B
2026.07
94.3
Stage1-SFT-v1
Params=4B
2026.07
90.9
Stage1-SFT-v0
Params=4B
2026.07
90.7
Stage1-SFT-v3
Params=4B
2026.07
90.2
Stage2-SFT
Params=4B
2026.07
90.1
YuFeng-XGuard-Reason-8B
Params=8B
2026.07
89.8
Stage1-SFT-v2
Params=4B
2026.07
88.9
YuFeng-XGuard-Reason-0.6B
Params=0.6B
2026.07
88.7
Qwen3Guard-4B-Gen
Params=4B
2026.07
85.1
Qwen3Guard-8B-Gen
Params=8B
2026.07
84.9
Stage1-SFT-v4
Params=4B
2026.07
84.2
Qwen3Guard-0.8B-Gen
Params=0.8B
2026.07
83.4
Feedback
Search any
task
Search any
task