421M-parameter open-source Laya returns answer probabilities in one pass at 40ms on a T4

maier_ak · x · 2026-09-28

Laya is a 421M-parameter open-source model (Apache 2.0) built by a solo developer: it takes text plus typed questions and returns answers with probabilities in a single forward pass—no text generation at all—running at 40 ms on one T4 GPU.

Its closed counterpart, Jev from TypeSafe AI, is API-only at $0.042 per million input tokens. Independent benchmarks put Jev at 264-276 ms median latency, 0.910 on AG News, and a calibration error of 0.246—the worst in that study.

Related event: Open-Source 421M Discriminative Model Laya Goes Viral: 7x Faster but Falters on Complex Tasks(5 posts)→

Original post →

More from Models

Models channel →