New API: The Next-Gen LLM Gateway & AI Asset Manager

New API: The Ultimate Open-Source LLM Gateway for Seamless AI Integration

In the rapidly evolving landscape of artificial intelligence, managing and integrating diverse large language models (LLMs) can be a significant challenge. Enter New API, an innovative open-source project designed to simplify this complexity. Building upon the foundation of One API, New API serves as a next-generation LLM gateway and AI asset management system, enabling users to unify various AI models, including OpenAI, Claude, Gemini, and DeepSeek, into a single, accessible format.

What is New API?

New API is a robust, multi-language supported platform engineered for both individual developers and enterprise environments. It provides a centralized system for managing and distributing AI model interfaces, making it incredibly easy to call different LLMs with a consistent API structure. This project is a powerful fork of the popular One API, enhancing its capabilities with significant new features and an improved user experience.

Key Features and Enhancements:

New API stands out with a rich set of functionalities tailored to meet modern AI integration needs:

• Unified API Access: Convert multiple large models into a single, standardized call format, supporting OpenAI, Claude, Gemini, and more.
• Revamped User Interface: A completely new, intuitive UI for enhanced user experience.
• Multi-language Support: Designed for a global audience with comprehensive multi-language capabilities.
• Online Top-up & Quota Management: Features online top-up options for easy payment and tools to query usage quotas by API key.
• Database Compatibility: Maintains compatibility with original One API databases for seamless migration.
• Flexible Billing: Supports per-call billing for models and channel weighted random distribution.
• Data Dashboard: An integrated console for monitoring usage and performance metrics.
• Advanced Controls: Implement token grouping, model restrictions, and advanced authorization methods (LinuxDO, Telegram, OIDC).
• Rerank Model Support: Seamless integration with Rerank models like Cohere and Jina for improved search and relevance.
• Realtime API & Claude Messages: Support for OpenAI Realtime API (including Azure channels) and Claude Messages format.
• Intelligent Reasoning Control: Configure reasoning effort for OpenAI (e.g., -high, -medium, -low suffixes) and enable 'thinking' mode for Claude models (-thinking suffix).
• Cache Billing: Option to charge for cached responses at a set ratio, optimizing cost efficiency.
• Extensive Model Support: Beyond major LLMs, it supports third-party models like Midjourney-Proxy (Plus) interface, Suno API, and custom channels.

Deployment Made Easy:

New API offers flexible deployment options to suit various environments:

• Docker Compose (Recommended): Get started quickly with a docker-compose.yml file.
• Direct Docker Image: Run New API effortlessly using pre-built Docker images, configurable with SQLite or MySQL databases.
• Baota Panel Integration: Simple one-click installation via Baota Panel's Docker feature.

For multi-machine deployments, crucial environment variables like SESSION_SECRET and CRYPTO_SECRET ensure consistent login states and Redis content access.

Caching and Reliability:

The system includes robust channel retry functionality and highly recommended caching features. By setting REDIS_CONN_STRING or enabling MEMORY_CACHE_ENABLED, users can significantly improve response times and operational efficiency.

Who Can Benefit?

New API is an invaluable tool for:

• Developers: Simplify the integration of diverse AI models into applications.
• Businesses: Efficiently manage AI resources, control access, and monitor usage across teams.
• Researchers: Experiment with various LLMs and API configurations from a unified platform.

New API represents a significant leap forward in AI model management, offering a comprehensive, flexible, and powerful open-source solution for anyone looking to harness the full potential of large language models. Explore its capabilities and join the growing community passionate about accessible AI.