Back to explore

lifesim-social-1 vs OpenAI Decisions : benchmark du modèle Jev

Cris Lenta publie des résultats de benchmark : lifesim-social-1 atteint 88 % de précision, 102 ms de latence p50 et 0,047 d'erreur de calibration sur les décisions sociales, surpassant l'API Decisions d'OpenAI.

we beat OpenAI's Decisions API 24h after it launched it's Jev competitor lifesim-social-1 vs OpenAI Decisions on social decisions: • accuracy: 88% vs 67% • p50 latency: 102 ms vs 278 ms • calibration error: 0.047 vs 0.170 (lower is better) this means agents will not only

lifesim-social-1 vs OpenAI Decisions : benchmark du modèle Jev video poster
· 8 likesOpen on X