Analysis updated 2026-05-18
Manage several upstream AI API relay admin panels from a single dashboard.
Track balance, cost, and usage trends across multiple upstream sites over time.
Run a scheduled pricing discount campaign that reverts automatically when it ends.
Get balance or pricing change alerts through DingTalk, Feishu, or Telegram.
| deviseo/transit-hub | internetkafe/deathcore | sphireinc/foundry | |
|---|---|---|---|
| Stars | 177 | 179 | 183 |
| Language | Go | Go | Go |
| Setup difficulty | hard | moderate | moderate |
| Complexity | 4/5 | 4/5 | 3/5 |
| Audience | ops devops | ops devops | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires PostgreSQL, Redis, and Docker Compose plus existing admin access to the upstream platforms you want to connect.
TransitHub is a self-hosted admin dashboard for people who run multiple upstream API relay services, the kind of sites that resell access to AI models like Claude, GPT, and Gemini through a shared endpoint. Instead of logging into each upstream admin panel separately, an operator connects them all to TransitHub and manages balances, pricing, and monitoring from one place. The project focuses on the everyday work of running these services. It lets an operator manage several admin workspaces at once, each with its own isolated data, add and sync upstream sites, and view live and historical dashboard metrics such as balances, costs, usage trends, and per group activity. It also tracks group pricing multipliers over time, including snapshots and a history of changes, and can run scheduled pricing campaigns, for example a temporary discount, that automatically restore the original pricing once the activity ends. There is also support for automatic pricing rules that react to changes coming from an upstream, and notification channels through DingTalk, Feishu, and Telegram bots for things like low balance warnings or pricing changes. Under the hood it is a Go backend paired with a Vue 3 frontend, using PostgreSQL for storage and Redis for caching and sessions. It is designed to be deployed with Docker Compose, and the project ships production compose files for that purpose. The README includes an upfront notice that this tool is meant for connecting to and operating upstream platforms an administrator already has legitimate access to, and it asks users to follow the terms of service of every upstream they connect, comply with local laws, and take responsibility for protecting their own credentials, backups, and secrets. It states the project is for technical learning and that the authors are not liable for service interruptions, account restrictions, or data loss resulting from its use. The README does not state a license, and it lists several third party API relay services as sponsors of the project.
A self-hosted admin dashboard for operators who run multiple upstream AI API relay services, unifying balances, pricing, and monitoring in one place.
Mainly Go. The stack also includes Go, Vue, PostgreSQL.
The README does not state a license, so default copyright rules apply and reuse permissions are unclear.
Setup difficulty is rated hard, with roughly 1h+ to a first successful run.
Mainly ops devops.
This repo across BitVibe Labs
Verify against the repo before relying on details.