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

jev-accounts-hub

Go gateway that pools multiple TypeSafe accounts behind the native System One API with user keys, quotas and failover.

ServicebetaMIT
View on GitHub ↗

Overview

jev-accounts-hub is a multi-account manager and API gateway for TypeSafe Jev, documented in Chinese. It pools several official TypeSafe accounts and handles authentication, quotas, usage tracking and failover in one place. Downstream callers keep using the official System One protocol (POST /v1/systemone and GET /v1/models) and only change the base URL; it is not an OpenAI chat-completions compatibility layer.

Features include weighted round-robin across accounts with automatic switching on 429, 529, 5xx or 401 responses, an outbound proxy pool (HTTP, HTTPS, SOCKS5) with optional per-account binding, user keys with RPM limits, token quotas, expiry and disabling, AES-256 encryption of upstream keys at rest, call logs, usage statistics, health checks and a playground. It runs with Docker Compose or Go 1.24, stores state in SQLite, and has an admin console with registration closed by default. It is MIT licensed.

What the decision model decides

TypeQuestionAnswers
OtherWhich pooled upstream account should serve this request, and should it fail over?(inferred)—

How it works

  1. Import TypeSafe account keys into the pool
  2. Issue user keys with limits to callers
  3. Callers point TYPESAFE_BASE_URL at the gateway
  4. Gateway authenticates, rate-limits and forwards to api.typesafe.ai
  5. Fail over to another account on errors or rate limits

Strengths

  • Official SDKs work by changing only the base URL
  • Encrypted storage of upstream keys
  • Per-user quotas, logs and usage stats

Caveats

  • Pooling accounts to bypass per-account limits may conflict with provider terms
  • Documentation is in Chinese

Getting started

`docker compose up --build` or `go run ./cmd/jevproxy`, then open http://127.0.0.1:8080.

At a glance

Type
Service
Status
beta
Language
Go
Runtime
HTTP gateway (Docker)
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

  • A multi-account manager and API gateway for TypeSafe / Jev. 一个用于 TypeSafe / Jev 的多账户管理器和 API 网关。交流群：1102910606.

    hellogumbo/awesome-jev

Frequently asked questions

What does the gateway do?

Pools multiple TypeSafe accounts behind one System One endpoint with user keys, quotas and failover.

Do clients need changes?

Only the base URL and a gateway-issued key.

Is it OpenAI-compatible?

No, it keeps the native /v1/systemone protocol.

Related projects

  • ai

    Hack Club's lightweight LLM proxy, listed for adding a Jev forwarding endpoint behind its auth, limits and usage logging.

    MCP serverproductionTypeScript
  • decision-gate

    npm library that rate-limits, spend-caps, caches and redacts every Jev request made in a loop.

    LibrarybetaJavaScript
  • jevkit

    Rust CLI for Jev that lints typed question sets offline, then sends them and prints probabilities, labels and scores as JSON.

    CLIbetaRust
  • jevtok-ts

    Offline TypeScript token counter and request-cost estimator for Jev 1.13, a port of the reconstructed jevtok encoder.

    PluginbetaTypeScript
  • new-api-typesafe-plugin

    Task plugin for the new-api LLM gateway that adds a native /v1/systemone endpoint for Jev with auth, routing and token billing.

    PluginbetaJavaScript
  • x-herald

    Chinese-documented LLM gateway with virtual model routing, protocol conversion and a transparent Jev System One proxy.

    ServicebetaTypeScript