Laya
AssessPlatforms
An open-source family of non-autoregressive decision models for fast structured prediction.
Why it's here
Placed in Assess: 3 article(s) of evidence from 1 source(s), led by open-source activity, with 3 in the last 30 days. Confidence 34%.
Evidence (3)
- 6Hacker News·9/25/2026open_sourceOllaya brings local decision models to Ollama-style workflows
Ollaya is an open-source project for running typed decision models locally on your own hardware. It exposes TypeSafe-compatible endpoints and supports open models such as Laya, Decider, NLI, and Gliclass for low-latency, calibrated answers to structured questions.
- 4Hacker News·9/20/2026researchLaya runs offline on Mac M4 with Core ML
A demo shows Laya (OS Jev) running offline on a Mac M4 using Core ML and producing about 45 decisions per second. The post highlights local inference performance on Apple Silicon rather than a broader product launch or model release.
- 6Hacker News·9/19/2026open_sourceOpen non-autoregressive decision models for structured tasks
The post describes Laya, an open-source family of non-autoregressive decision models designed to make fast, calibrated structured predictions from text and other inputs. It positions the models as a System 1 alternative to large generative LLMs for tasks like routing, classification, and scoring, and claims support for more than 100 languages with Apache 2.0 weights.