GPT-Load Logo

GPT-Load

High-Performance AI API Transparent Proxy

Enterprise-grade AI API proxy service developed with Go 1.23+, supporting multiple AI providers like OpenAI, Google Gemini, Anthropic Claude. Features intelligent key management, load balancing, high concurrency handling, and comprehensive monitoring.

DocumentationInstallationGitHub
Go 1.23+
Backend Language
MySQL
Data Storage
Redis
Cache System
Vue 3
Admin Interface
Quick Start
docker run -d --name gpt-load \
    -p 3001:3001 \
    -e AUTH_KEY=your-secure-key-here \
    -v "$(pwd)/data":/app/data \
    ghcr.io/tbphp/gpt-load:latest

# 访问管理界面：http://localhost:3001
⚠️
Security Warning
Please replace 'your-secure-key-here' with a strong key! Using default or simple keys poses serious security risks.

System Architecture

GPT-Load adopts a three-tier architecture design to provide high-performance, high-availability AI API proxy services

Data Flow Architecture

Client Applications

Web/Mobile apps call through standard OpenAI API format

HTTP/HTTPS Requests
Bearer Token Auth
JSON Format Interaction

GPT-Load Proxy Layer

Core proxy service responsible for request forwarding and management

Transparent Proxy
Key Management
Load Balancing
Request Logging

AI Service Providers

Unified access to multiple AI services

OpenAI API
Google Gemini
Anthropic Claude
Other Compatible Services

Infrastructure Components

MySQL 8.2+

Persistent Storage

Config Data
User Info
Request Logs

Redis

Cache & Locks

Key Cache
Distributed Locks
Session Storage

Management Interface

Web Control Panel

Visual Config
Monitor Dashboard
Log Viewer

Flexible Deployment Options

Standalone Deployment

  • • Docker Compose one-click startup
  • • Includes complete MySQL + Redis
  • • Suitable for development and small production

Cluster Deployment

  • • Master/Slave architecture
  • • Horizontal scaling support
  • • High availability guarantee
Quick Start

Start GPT-Load in 3 Steps

Quick deployment via Docker Compose, including complete database and cache services

1. Clone Project

Download complete project code from GitHub

git clone https://github.com/tbphp/gpt-load.git
cd gpt-load

2. Configure Environment

Copy and edit environment configuration file

# Copy environment configuration file
cp .env.example .env

# Edit configuration (optional)
# vim .env

# Main configuration items:
# APP_PORT=3001
# APP_SECRET=your-secret-key

3. Start Services

Use Docker Compose for one-click startup

# Start services (including MySQL and Redis)
docker compose up -d

# Access admin interface
# http://localhost:3001

System Requirements

Go 1.23+
Runtime Environment
MySQL 8.2+
Data Storage
Redis
Cache Service
Docker
Container Runtime
View Detailed Deployment Documentation

Start Using GPT-Load Now

Deploy in minutes and enjoy high-performance AI API proxy services

Quick DeployView DocsGitHubSponsor