Laya Python Packages

prompture

Prompture is an API-first library for requesting structured JSON output from LLMs (or any structure), validating it against a schema, and running comparative tests between models.

laya-mlx

Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.

datenwissenschaften

Training engine where the Laya decision model learns to beat retro games: text game state, group-relative policy gradients, per-state checkpoints and a reverse curriculum.

laya-apple

Correctness-validated heterogeneous Laya runtime for Apple Silicon (MLX GPU + Apple Neural Engine)

laya-coreml

Local Laya typed decisions on Apple Core ML and Neural Engine. Validated ports, ~5 ms short decisions on M3 Max, reproducible speed and energy benchmarks.

fastfence

A policy gateway for agent tools, resources, and local LLMs

luar

LUAR (Local Utility for Automated Reviews): drop a spreadsheet, ask choice/yes-no questions, get answers per row from a local decision model (Laya). No API keys, nothing leaves your machine.

edgejev

离线可用的本地类型化决策：4 核 CPU 单题 15.6ms。Local & offline Jev / System One inference on CPU — ONNX + INT8, no torch at runtime. 支持 laya / kev / PlayJev

entune

Open-source, cross-platform dictation with your choice of speech model and a personal dictionary that selects corrections in context. 🌟 Star to support our work!

l2s1-sdk

jev-style adaptor for llama.cpp interface

deqio

Run fast, typed AI decision models behind one API. Choice, score and yes/no decisions with multiple engines and hardware backends.

pydecide

One client for every decision model: Choice, Score and Noul over Jev, OpenRouter, laya, MLX, CrossEncoders and LLM fallback

laya-mcp

MCP server for Laya typed decisions (noul / choice / score): warm model sidecar, token-budget preflight, persisted calibration, honest confidence.

stuntd

Local proxy that learns your app's typed LLM decisions and answers them with a Laya head. Jev and OpenAI compatible.

watfile

Text/PDF - File categorization and sorting with Typesafe AI Jev or local calibrated decision model

system-one

Vendor-neutral SDK for typed decision-making (yes/no, choice, score) — hosted and local

leanroute

A fast decision layer in front of any LLM: blocks attacks, sends easy requests to cheap models, hard ones to strong models, and shows the money saved. OpenAI-compatible, self-hostable, powered by Laya.

laya-finetune

Fine-tune Laya, the open System One decision model, for any typed-decision task: synthetic data, RLCD, calibration and long context.

ruhui

Multilingual, non-autoregressive System 1 decision model.

sys1bench

Benchmark harness for typed System One decision models (choice / score / noul): calibration vs noise floor, framing sensitivity, selective prediction, ordinal fidelity, interference, robustness.

laya-cuda

A lightweight CUDA inference library for Laya

mcp-laya

Safe-by-default MCP server for Laya — fast, local, non-autoregressive typed decisions (classify / score / yes-no) for AI agents, in 100+ languages.

laya-decision-api

把本地 System-1 决策模型（Laya）包成 HTTP 服务：意图分类 / 工单分派 / LLM 路由 / 内容审核，离线零 token、不生成文字；Apple Silicon(MLX) 与 Linux(torch) 双后端；可导出 OpenAPI 3.0/3.1；Java/TS/Python 零依赖 SDK。

laya-phishield

Explainable phishing email detection: atomic semantic signals from a local System 1 decision model plus deterministic header/URL checks, combined by logistic regression