new-api MCP Server: LLM Gateway for Unified AI Model Access
new-api MCP Server: LLM Gateway for Unified AI Model Access
new-api serves as an MCP Server that functions as a next-generation LLM gateway and AI asset management system, built in JavaScript. It provides a unified interface for calling various large language models, including OpenAI and Claude, making it a practical solution for individuals or enterprises needing to manage and distribute internal AI channels. The project is a secondary development based on One API.
Unified LLM Interface
The core capability of new-api is its ability to transform diverse large language models into a consistent format for invocation. This means that whether you're working with OpenAI or Claude, new-api standardizes the API calls, simplifying integration and management. This abstraction layer is particularly useful for organizations that utilize multiple LLM providers and need a single point of control.
Deployment and Management
For deployment, new-api offers Docker images available on both GHCR.io and Docker Hub. This containerized approach streamlines setup and ensures consistent environments across different deployments. The project also provides a clear licensing under MIT, indicating its open-source nature and suitability for personal learning and internal enterprise use. It's important to note the project explicitly states it's for personal learning and does not guarantee stability or provide technical support.
Project Origin and Status
new-api is a fork and secondary development of the One API project. It maintains an active development presence, as indicated by its GitHub repository with over 51,540 stars. The project's badges also show current release versions and GoReportCard status, highlighting ongoing maintenance and code quality efforts.