Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

SLIMORCA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Utility EvaluationSLIMORCA (test)
Score68.85
24
Win rate evaluationSLIMORCA (test)
Win Rate88.12
8
Language ModelingSlimOrca (test)
Test PPL3.81
3
Safety EvaluationSLIMORCA OOD Safety Scenario (test)
Harmfulness Score2.4
2
Showing 4 of 4 rows