About this project

AI Gateway is a high-performance, scalable AI model API gateway that provides a unified entry point for accessing various large model services. It is compatible with the OpenAI API format, allowing developers to integrate multiple providers such as OpenAI, Anthropic (Claude), DashScope, and Ollama using a single set of code. It supports full-modal functions including Chat, Vision, Image Generation, Speech-to-Text, Text-to-Speech, and Embeddings. Built on Go + Gin, the system possesses high concurrency processing capabilities, and its stateless design supports horizontal scaling. It features built-in Prometheus metrics, structured logging, request tracing, token bucket rate limiting, and API Key quota management, supporting MySQL/PostgreSQL storage backends and memory/Redis caching. It provides health checks, API Key management, and usage statistics, making it easy for enterprises or individuals to deploy private AI services. The project structure is clear, allowing developers to easily add new Providers. It is suitable for scenarios requiring unified access to multiple AI services with traffic control, monitoring, and cost management.