# GPT-Load

> An open-source, self-hosted AI gateway routing one endpoint to 35 upstream channels — key credentials and subscription logins, with failover, logs and cost estimates.

- **Category:** API Gateway
- **Licence:** MIT
- **Latest version:** v2.0.0-rc.57
- **Open source:** oss
- **Languages:** Typescript, Go, Vue
- **Platforms:** Docker Compose, Linux, Windows
- **Tags:** OAuth, LLM, Anthropic, Claude, Gemini, Openai, Api Gateway, Ai Gateway, Load Balancer, Go, Golang, Failover, Self Hosted, Codex, Claude code, Usage tracking, Openai compatible, Llm gateway, Anthropic api, Gemini Api, Gin, API Key Pool, Subscription Accounts, Cost Estimates, Request Logs

## Description

GPT-Load is an open-source gateway for AI traffic that you run on your own hardware: applications get a single base URL plus an issued AccessKey, and the service spreads requests over many providers and credentials. Client-side it speaks the native OpenAI Chat Completions and Responses, Anthropic Messages, and Gemini protocols (also images, embeddings, rerank, and some Mistral routes), routing out to roughly thirty-five built-in channels: official vendor APIs, cloud offerings such as AWS Bedrock, Azure OpenAI and Vertex AI, model services, subscription accounts (Codex, Claude, Antigravity, Grok) and any OpenAI-compatible relay. Scheduling is the main job — weights across credentials, retries, per-model cooldowns, blacklisting with recovery probes, session affinity, and a route-check tool in the console that explains why a call failed. The management UI is baked into the same binary and covers groups, model aliases, AccessKeys with rate or cost limits, request logs, credential health and per-model cost estimates, while upstream secrets are encrypted locally. Installation is the official Compose file, a portable native binary for Linux, macOS or Windows, or a Windows service installer; storage starts on a managed SQLite file and can point to MySQL or PostgreSQL. Caveats: the 2.0 line is a pre-GA rewrite that cannot read or import 1.x data, it is designed for one instance only, and cost figures are operational estimates rather than invoices.

## Links

- [Homepage](https://github.com/tbphp/gpt-load)
- [Website](https://www.gpt-load.com)
- [Docker](ghcr.io/tbphp/gpt-load)
- [GitHub repository](https://github.com/tbphp/gpt-load)
- [1.4.x documentation archive](https://www.gpt-load.com/v1)

---
Source: https://selfhost.directory/project/gpt-load