Announcement

Free to view yesterday and today
Customer Service: cat_manager

One-API: Unified LLM API Management & Key Redistribution System

A unified LLM API management and key redistribution system that supports multiple mainstream large language models, providing a single API endpoint for simplified access and management. It comes as a single executable and is Docker-ready for easy, one-click deployment.

JavaScript
Added on 2025年6月21日
View on GitHub
One-API: Unified LLM API Management & Key Redistribution System preview
25,790
Stars
5,241
Forks
JavaScript
Language

Project Introduction

Summary

This project is an LLM API management and distribution system designed to unify access to various large language models through a single API interface. It supports a wide range of providers and includes features for key management and redistribution, packaged for easy deployment.

Problem Solved

Managing access to multiple LLM providers with different APIs and keys is complex and inefficient. This project solves this by providing a single, unified gateway that simplifies integration, manages keys centrally, and facilitates redistribution.

Core Features

Unified API Endpoint

Consolidates APIs from OpenAI, Azure, Anthropic, Google, DeepSeek, Doubao, ChatGLM, Wenxin Qianfan, Xunfei Xinghuo, Tongyi Qianwen, 360 Zhinao, Tencent Hunyuan, and more under a single, consistent API interface.

Centralized Key Management

Provides robust key management capabilities, including usage tracking and rate limits.

API Key Redistribution

Allows for secondary distribution or resale of API access.

Easy Deployment (Single Binary & Docker)

Simplified deployment via a single executable file and official Docker image.

Intuitive User Interface

Offers a user-friendly interface for configuration and monitoring.

Tech Stack

Go
Docker

Use Cases

The system is applicable in various scenarios where simplified and centralized management of diverse LLM APIs is required.

Scenario 1: Centralized API Access for Organizations

Details

A company requires access to multiple LLM providers (e.g., OpenAI for general tasks, Claude for creative writing, DeepSeek for coding). Instead of integrating with each API separately, they use ONE-API as a single gateway.

User Value

Simplifies development and maintenance, ensures consistent access control and usage monitoring across all models.

Scenario 2: Simplifying Developer Integrations

Details

A developer building an application wants to easily switch between different LLM backends or use the best model for a specific task without changing application code.

User Value

Reduces integration complexity, allows for easy experimentation and dynamic switching between models.

Scenario 3: Managing and Redistributing API Keys

Details

A team or platform wants to manage and share a pool of LLM API keys among multiple users or projects, potentially tracking individual usage.

User Value

Provides a mechanism for secure key sharing, usage tracking, and potential cost allocation or redistribution models.

Recommended Projects

You might be interested in these projects

charmbraceletbubbletea

Bubble Tea is a simple yet powerful framework for building terminal user interfaces (TUIs) in Go, based on the The Elm Architecture.

Go
31840911
View Details

mywalkbLSPosed_mod

This project is a modified version (fork) of the original LSPosed framework, aiming to provide enhanced features, performance optimizations, and potentially broader compatibility for advanced Android users and module developers.

Java
4193232
View Details

hexgradkokoro

Kokoro-82M is a compact and efficient AI model optimized for various text generation tasks, making it ideal for applications requiring fast inference and lower resource usage.

JavaScript
3043329
View Details