fruitbars/simple-one-api
OpenAI 接口接入适配，支持千帆大模型平台、讯飞星火大模型、腾讯混元以及MiniMax、Deep-Seek，等兼容OpenAI接口，仅单可执行文件，配置超级简单，一键部署，开箱即用. Seamlessly integrate with OpenAI and compatible APIs using a single executable for quick setup and deployment.
What it solves
It provides a unified gateway to connect multiple large language model (LLM) providers, eliminating the need to manage multiple different API protocols. It allows users to interact with various models through a single interface while supporting standard client protocols like OpenAI Chat Completions, OpenAI Responses, and Anthropic Messages.
How it works
The project acts as a proxy server that routes requests to various AI providers. It features a routing system that can distribute requests using random, preferred, polling, or hash-based methods. It includes a built-in React-based web chat interface and a visual configuration console for managing providers, models, and credentials. The system uses SQLite for configuration storage and can be deployed as a standalone server or a Wails-based desktop application.
Who it’s for
It is designed for users and developers who want a centralized way to manage multiple AI model providers and access them via a single API gateway, as well as those who prefer a visual interface for managing their AI services.
Highlights
- Multi-protocol support: Compatible with OpenAI and Anthropic API formats.
- Flexible routing: Supports random, preferred, polling, and hash routing across multiple providers and credentials.
- Integrated UI: Includes a built-in web chat interface and a visual configuration dashboard.
- Resilience features: Implements circuit breaking and half-open recovery at the provider and model level.
- Deployment options: Available as a single-file server, Docker image, or a desktop application.
Related
- Project
- Project
- Project
- Project
- Project