One API: Your Universal Interface for LLMs, Empowered by UBOS

In the rapidly evolving landscape of Large Language Models (LLMs), managing diverse APIs and integrating them into your applications can be a daunting task. One API emerges as a powerful solution, offering a unified interface to access a multitude of LLMs through a standardized OpenAI-compatible API format. This eliminates the complexities of dealing with individual API implementations, streamlining your AI development workflow.

Built for the Future of AI by UBOS: The AI Agent Development Platform

One API seamlessly integrates with the UBOS platform, a full-stack AI Agent development environment. UBOS is focused on empowering businesses by bringing AI Agents to every department. Our platform allows you to orchestrate AI Agents, connect them to your enterprise data, and build custom AI Agents using your preferred LLM models and Multi-Agent Systems. With the combined power of One API and UBOS, you can unlock unprecedented levels of efficiency and innovation in your AI initiatives.

Key Features

• Unified API Access: Interact with a wide range of LLMs using a single, consistent OpenAI-style API.
• Broad Model Support: Supports leading LLMs including OpenAI (ChatGPT series, Azure OpenAI API), Anthropic Claude (AWS Claude), Google PaLM2/Gemini, Mistral, 字节跳动豆包, 百度文心一言, 阿里通义千问, 讯飞星火认知, 智谱 ChatGLM, 360 智脑, 腾讯混元, Moonshot AI, 百川大模型, MINIMAX, Groq, Ollama, 零一万物, 阶跃星辰, Coze, Cohere, DeepSeek, Cloudflare Workers AI, DeepL, together.ai, novita.ai and 硅基流动 SiliconCloud. This extensive support future-proofs your applications as new models emerge.
• Flexible Deployment: Deploy using Docker, Docker Compose, or manually. One-click deployment options are available on platforms like Sealos and Zeabur. Works out-of-the-box.
• Token Management: Implement fine-grained control over API usage with token-based access control. Set expiration times, usage quotas, IP restrictions, and allowed model access for each token. This is crucial for managing costs and ensuring security.
• Channel Management: Batch create and manage channels for different LLM providers. Configure model lists for each channel to optimize performance and cost.
• User & Channel Grouping: Organize users and channels into groups and apply different usage multipliers for flexible pricing and resource allocation.
• Rate Limiting: Protect your system and prevent abuse with configurable global API and web request rate limits.
• Load Balancing: Distribute API requests across multiple channels for improved performance and resilience. This ensures high availability even when some LLM providers experience outages.
• Stream Mode Support: Enable real-time, typewriter-style output using stream mode for a more engaging user experience.
• Multi-Machine Deployment: Scale your One API deployment across multiple servers for high availability and increased capacity.
• Quota Management: Track and manage API usage with detailed quota breakdowns.
• Redemption Code Management: Issue and manage redemption codes to allow users to replenish their accounts.
• Referral Rewards: Incentivize user growth with referral reward programs.
• Model Mapping: Redirect user requests to specific models for customized deployments. This is useful for testing new models or optimizing performance.
• Automatic Retries: Automatically retry failed API requests to improve reliability.
• Image Generation: Supports image generation APIs, expanding the range of AI-powered applications you can build.
• Cloudflare AI Gateway Integration: Seamlessly integrate with Cloudflare AI Gateway for enhanced security and performance.
• Customization: Tailor the look and feel of your One API instance with custom system names, logos, footers, home pages, and about pages. Use HTML and Markdown or embed external web pages using iframes.
• Extensible API: Expand and customize One API functionality without modifying the core codebase using the system access token-based management API.
• User Authentication: Supports multiple user login and registration methods, including email (with whitelisting), Feishu (Lark), GitHub OAuth, and WeChat OAuth (requires a separate WeChat Server deployment).
• Theme Switching: Customize the user interface with different themes.
• Message Pusher Integration: Integrate with Message Pusher to send alerts and notifications to various apps.

Use Cases

• AI-Powered Applications: Build AI-powered applications that leverage the best LLMs without being tied to a single provider.
• Chatbots & Virtual Assistants: Create intelligent chatbots and virtual assistants that can understand and respond to user queries with human-like accuracy. UBOS provides the tools to fine-tune these agents with enterprise data.
• Content Generation: Automate content creation tasks such as writing articles, generating marketing copy, and creating social media posts.
• Code Generation: Assist developers with code generation, debugging, and documentation.
• Data Analysis: Extract insights from large datasets using LLMs for natural language processing and data analysis.
• Educational Tools: Develop interactive learning tools that provide personalized feedback and guidance.
• Research & Development: Experiment with different LLMs and compare their performance for research purposes.
• Enterprise AI Infrastructure: Build a robust and scalable AI infrastructure for your organization, enabling various departments to leverage AI capabilities.

Installation and Deployment

One API offers flexible deployment options to suit your infrastructure and needs:

• Docker: The recommended method for easy and consistent deployment.
• Docker Compose: For orchestrating multi-container deployments.
• Manual Deployment: For advanced users who prefer to build and run the application manually.

Detailed deployment instructions are available in the project’s README.

Configuration

One API is designed to be highly configurable, allowing you to tailor it to your specific requirements. Configuration can be done through environment variables or command-line parameters.

Key Configuration Options:

• Database: Choose between SQLite (default) and MySQL for persistent data storage. MySQL is recommended for high-concurrency environments.
• Caching: Enable Redis caching to improve performance and reduce database load.
• Rate Limiting: Configure global API and web request rate limits to protect your system.
• Authentication: Configure user authentication methods and settings.
• Themes: Customize the look and feel of the application.

How it Works

One API acts as a central hub for routing API requests to different LLM providers. When a user sends a request to One API, it forwards the request to the appropriate LLM provider based on the configured channel and load balancing settings. One API then relays the response back to the user.

Common Issues and Solutions

The project’s README includes a comprehensive FAQ section that addresses common issues and provides solutions.

Related Projects

• FastGPT: A knowledge-based question answering system based on LLMs.
• ChatGPT Next Web: A cross-platform ChatGPT application.
• VChart: A multi-terminal chart library.
• VMind: An intelligent visualization solution.

License

One API is licensed under the MIT License. You must retain the attribution and link back to the project in the page footer. Contact the developers for licensing options if you do not wish to retain the attribution.

Unlock the Potential of LLMs with One API and UBOS.

By unifying access to diverse LLMs and integrating seamlessly with the UBOS platform, One API empowers you to build innovative AI-powered applications with unprecedented ease and scalability. Take control of your AI development workflow and drive your business forward with the power of One API and UBOS.