Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

laya-server

Self-hosted Docker service and web console that serves Laya's multilingual model on Jev's /v1/systemone format behind API keys.

Servicebeta
View on GitHub ↗

Overview

LAYA SERVER, from the 1Panel team, packages upstream Laya's System One inference as a standalone, self-hosted service with a web interface, compatible with TypeSafe's Jev API format. You run one Docker container with an admin username and password, sign in to the console, and create API keys that are shown once. Applications then call POST /v1/systemone with a bearer key and receive answers, the model name and actual token usage.

The published image includes the multilingual model, selectable with model set to auto or multilingual. A readiness endpoint checks model files and data persists in a volume. Usage statistics are written asynchronously in batches to SQLite without delaying responses; the README describes them as best-effort, eventually consistent telemetry rather than a billing or quota ledger. Documentation is available in English and Chinese. It suits teams that want a managed-feeling local decision API with key management.

What the decision model decides

TypeQuestionAnswers
Yes / noDoes the customer ask for a refund?—
ChoiceAny choice question via the Jev-format API—

How it works

  1. docker run the image with admin credentials
  2. Sign in and create an API key
  3. Call /v1/systemone with the key
  4. View usage statistics in the console

Strengths

  • One-container deployment
  • API key management and console
  • Jev-compatible request format

Caveats

  • Statistics are not a billing ledger
  • Only the multilingual model is in the published image
  • HTTPS must be set up via a reverse proxy

Getting started

docker run -d --name laya-server -p 8080:8080 -v laya-data:/data -e LAYA_ADMIN_USERNAME=admin -e LAYA_ADMIN_PASSWORD=... 1panel/laya-server:latest

At a glance

Type
Service
Status
beta
Language
Python
Runtime
Docker service with web console
Has code
yes

Use cases and integrations

How curated lists describe it

  • A self-hosted API and web interface for Laya.

    wh000wh000/awesome-jev-live

  • A self-hosted API and web interface for Laya’s structured decision models, compatible with the TypeSafe Jev API format.

    hellogumbo/awesome-jev

  • Self-hosted API and web interface for Laya's checkpoints, speaking `/v1/systemone` behind its own API keys.

    MrJev/awesome-jev

Frequently asked questions

Is it compatible with Jev clients?

Yes, it accepts the Jev /v1/systemone format with bearer keys.

Which model is included?

The multilingual Laya model.

Can I bill users from its statistics?

The README says statistics are best-effort telemetry, not a billing ledger.

Related projects

  • decidealot

    Docker service exposing local Laya, Von, a CLM head or hosted Jev through one TypeSafe-compatible HTTP and MCP endpoint.

    MCP serverbetaPython
  • edgejev

    Converts open Jev reproductions (Laya, Kev, NanoJev, PlayJev) to INT8 ONNX for offline CPU decisions at about 15.6 ms per question.

    ServicebetaPython
  • go-system-one

    Native Go runtime serving Gemma 4 12B as a Jev-like decision model with hand-tuned PTX kernels and no CGo or CUDA toolkit.

    ProjectbetaGo
  • gojev

    Go harness for System One models: hosted Jev, open Kev in-process via llama.cpp, and Cloudflare Clef, with typed helpers and a CLI.

    ServicebetaGo
  • hearim

    Go gateway exposing the Jev /v1/systemone contract over ordinary LLM backends by reading single-token label logprobs.

    ServicebetaGo
  • intern-decision-mlx

    Runs Intern-Decision-0.8B on Apple Silicon behind a local /v1/systemone endpoint that accepts screenshots for agent decisions.

    ServicebetaPython