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

BrighTO Router

Self-hosted Rust LLM gateway that puts OpenAI, Anthropic and Jev-style /v1/systemone backends behind one endpoint with load balancing.

Servicebeta
View on GitHub ↗

Overview

BrighTO-Router is an open-source, self-hosted LLM gateway, model load balancer and System One decision router written in Rust and installed with Docker. It gives a team one stable endpoint for OpenAI-compatible, Anthropic-compatible, cloud and local backends, keeps provider keys private, records usage in PostgreSQL and offers NGINX-style model groups with round-robin or weighted load balancing, fallback routing and token budgets. It positions itself as a narrower alternative to LiteLLM or Bifrost.

For decision models, it exposes /v1/systemone and /v1/decisions routes for TypeSafe/Jev-compatible models such as Ollaya/Laya or hosted Jev-style endpoints, with an Ollaya System One provider preset in its portal. It also ships a single-file Python SDK and docs for agent clients.

What the decision model decides

TypeQuestionAnswers
OtherForwards Jev-style decision requests to configured backends; decisions are the caller's(inferred)—

How it works

  1. Configure providers and model groups in the portal
  2. Clients call one endpoint (chat or /v1/systemone)
  3. Gateway load-balances and falls back across backends
  4. Usage is recorded in PostgreSQL

Strengths

  • One endpoint for chat and decision models
  • Keys stay on the server
  • Weighted load balancing and budgets

Caveats

  • Self-hosting and Postgres required
  • Decision quality depends on the backend

Getting started

Follow the Docker quick start in the README.

At a glance

Type
Service
Status
beta
Language
Rust
Runtime
self-hosted gateway (Docker + PostgreSQL)
Has code
yes

Use cases and integrations

How curated lists describe it

  • BrighTO LLM Router: free open-source, ultra-fast Rust LLM gateway with a…

    wh000wh000/awesome-jev-live

  • Self-hosted Rust gateway that fronts OpenAI-, Anthropic- and `/v1/systemone`-shaped backends behind one endpoint, with weighted model groups and usage recorded in PostgreSQL.

    MrJev/awesome-jev

Frequently asked questions

Does it support Jev-style APIs?

Yes, /v1/systemone and /v1/decisions routes.

Can I use local decision models?

Yes, for example Ollaya/Laya.

Where is usage stored?

In PostgreSQL.

Related projects

  • s1-auto-router

    Claude Code plugin where a local Decider model on your Mac picks the model and effort level for each subagent task.

    PluginbetaPython
  • HiRoute

    Local-first routing engine for long-running agent tasks that uses a Jev decider to pick a model group per turn.

    PluginproductionRust, TypeScript
  • LiteLLM

    Open-source AI gateway for 100+ LLMs that can use Jev to classify requests for complexity routing and to compact tool results.

    ProxyproductionPython
  • openchamber

    Open-source workspace for running and reviewing AI coding agents across devices, with an optional Jev-based automatic model router.

    ProjectproductionTypeScript
  • jev-router★ 593

    Per-turn model routing for Claude Code and Codex: wraps the real CLI and lets Jev send simple turns to fast models and hard ones to strong.

    ProxybetaJavaScript
  • jev-codex-router★ 274

    Per-call router for Codex where Jev picks the model and thinking effort for every model call, including tool continuations.

    ProjectbetaPython