Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Win rate evaluation on SLIMORCA (test)
Loading...
88.12
Win Rate
TOSS
48.4752
58.7676
69.06
79.3524
Mar 1, 2026
Win Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Win Rate
TOSS
Backbone=Llama-3-8B-In...
2026.03
88.12
TOSS
Comparison model=Stand...
2026.03
68.37
TOSS
Paradigm=Training-base...
2026.03
68.37
TOSS
Comparison model=SEAL,...
2026.03
63.96
Circuit Breakers
Paradigm=Training-base...
2026.03
50.95
SafetyLock
Paradigm=Training-free...
2026.03
50.72
Standard SFT
Backbone=Llama-3-8B-In...
2026.03
50
Standard SFT
Paradigm=/, Base Model...
2026.03
50
Feedback
Search any
task
Search any
task