Back to explore

Jev Skill Suggestion benchmarkat mot Claude-modeller

Jev Skill Suggestion benchmarkades mot Claude-modeller med 26 skills, 10 prompts och 3 körningar vardera: Jev 90 % noggrannhet vid 330 ms median; Haiku 80 % vid 1,24 s; Sonnet/Opus/Fable 100 % vid 2,2–4,5 s. Jev byter lite noggrannhet mot 4–14× snabbare beslut.

I benchmarked Jev Skill Suggestion against Claude models with 26 skills, 10 prompts, and 3 runs each - Jev: 90% accuracy, 330ms median - Haiku: 80%, 1.24s - Sonnet/Opus/Fable: 100%, 2.2–4.5s So Jev trades a bit of accuracy for a 4–14× faster decision, while keeping the skill

Jev Skill Suggestion benchmarkat mot Claude-modeller 1
· 67 likesOpen on X