Laya: Open-Weight System 1 Decision Model Answers in 33ms per Pass
Balance- · reddit · 2026-09-21
Convai Innovations released Laya, an Apache 2.0 'System 1' decision model and open alternative to TypeSafe's closed Jev API. Instead of generating text, it takes a state (text, email, ticket, or JSON) plus typed questions (choice, score, noul) and answers all in one forward pass in 33–40ms on GPU — no parsing, nothing to hallucinate.
- Three checkpoints: a 421M English model on ModernBERT-large, a faster 322M multilingual model on mmBERT-base covering 100+ languages, and a typed-decisions-tuned variant, with a built-in Router detecting input script.
- Trained with RLCD using strictly proper scoring rules so honest probabilities maximize reward; includes an act-vs-escalate head for human handoff.
- Author reports beating Jev on AG News, emotion classification, and typed-decisions while running 6–8× faster (Jev figures are third-party numbers).
- Candid limits: near-chance without fine-tuning on typed-decisions; sharp drops beyond 50 options (Banking77: 0.425 vs Jev's 0.870); ordinal scoring weakest; fails on non-Latin scripts; ships overconfident — fit a temperature on your own data first.
Related event: Open-source Laya emerges hours after Jev goes closed-source(4 posts)→
More from Models
- Why Is Inkling 2 Small Trained on So Few Tokens Yet Still Competitive? — georgejrjrjr · 2026-11-12
- OpenAI's five biggest failures laid out: pretraining decline, alignment fears, enterprise lost to Claude — teortaxesTex · 2026-11-12
- Fable 5.2 Hits 41.3% on Remote Labor Index, Up From 2.5% a Year Ago — alliekmiller · 2026-11-12
- Beaker automates MIT's weight-space programming: one-neuron edit lifts MATH-500 from 7% to 37% for under $4 — VictorSanh · 2026-11-12
- Researcher says Anthropic dodged questions on Mythos false positive rate; 37,966 findings, 2 exploited — OwariDa · 2026-11-12
- Grok spotted building meeting "Listen to Conversation" feature with live transcript — nima_owji · 2026-11-12