QuantumNous/new-api

New API is an AI gateway and model management system that unifies access to multiple large language models through format conversion and centralized distribution.

The tool solves the problem of managing diverse AI model APIs by providing a single gateway that converts various LLMs into OpenAI-compatible, Claude-compatible, or Gemini-compatible formats. This approach allows organizations to abstract away differences between upstream providers, enabling applications to work with multiple models through standardized interfaces. The system functions as a centralized hub for aggregating different model services and distributing them to clients, while also providing organization-level authentication, usage analytics, and cost accounting capabilities.

Organizations should adopt this tool if they need to manage multiple AI model subscriptions across different providers, want to standardize API interactions across their applications, or require centralized governance and billing for AI services. It suits enterprise deployments and personal use cases where consolidating access to models like OpenAI, Claude, Deepseek, and Gemini behind a single gateway reduces integration complexity. The tool is designed for private deployment scenarios where users maintain control over their infrastructure and API keys.

The project maintains active development with regular updates and Docker deployment support. The codebase is written in Go, suggesting a focus on performance and efficient resource usage for gateway operations. The tool includes comprehensive documentation available in multiple languages, indicating attention to accessibility for international users. The project emphasizes lawful use and compliance, with explicit guidance that users must obtain upstream API keys legitimately and comply with applicable terms of service and regulations.