What it does The package runs Convai Innovations’ Laya model through ONNX Runtime for typed classification and scoring in Node.js and TypeScript.
Evidence 1 ↗
How Jev fits It exposes a request and response shape compatible with Jev’s system_one API while running Laya weights locally.
Evidence 1 ↗
Getting started Install @receptron/laya on Node.js 20 or newer, load Laya, then pass state and typed questions to systemOne; close the instance afterward.
Evidence 1 ↗ · Evidence 2 ↗
Scope and limits Question options must fit 192 tokens. The English checkpoint truncates state at its 512-token limit after the question header, and numeric JSON formatting can differ from Python.
Evidence 1 ↗
Evidence checked: Sep 30, 2026
Project facts
Relationship to Jev Local alternative
Evidence Documented
Language TypeScript
License MIT
Origin Original repository
Repository status Not archived
Created 2026-09-19
Last code update 2026-09-21
GitHub stars 589
Evidence checked Sep 30, 2026
Metadata checked Sep 29, 2026
Check status Up to date Stars measure the whole repository, including work unrelated to Jev.
Evidence and scope Documented records the linked documentation or source. JevHunt has not independently run or benchmarked this project.
Evidence commit: 6478649e723122ca24bbf5fb69ed1010023c9750
Discovered through: typesafe-field-guide.
How we review →