fx's safety classifier benchmarked: ~5-18x faster and more accurate than GPT-5.6-Luna

cramforce · x · 2026-09-17

fazxes benchmarked fx auto mode's safety classifier against typesafeai's Jev, finding it 5-18x faster and more accurate than GPT-5.6-Luna, their previous top choice.

Related event: Vercel fx to adopt Jev safety classifier, 5-18x faster than GPT Luna(2 posts)→

Original post →

More from Models

Models channel →