Back to explore

Jev Skill Suggestion comparado con modelos Claude

Se comparó Jev Skill Suggestion con modelos Claude usando 26 habilidades, 10 prompts y 3 ejecuciones cada uno: Jev 90% de precisión y 330 ms de mediana; Haiku 80% y 1,24 s; Sonnet/Opus/Fable 100% y 2,2–4,5 s. Jev sacrifica algo de precisión por decisiones 4–14× más rápidas.

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 comparado con modelos Claude 1
· 67 likesOpen on X