AIClient2API

AIClient2API is an API proxy service that breaks through client limitations. It converts free large models such as Gemini, Antigravity, Qwen Code, and Kiro, which were originally limited to use within the client, into a standard OpenAI-compatible interface that can be called by any application. Built based on Node.js, it supports intelligent interoperability of the three major protocols OpenAI, Claude, and Gemini, allowing tools such as Cherry-Studio, NextChat, and Cline to use advanced models such as Claude Opus 4.5, Gemini 3.0 Pro, and Qwen3 Coder Plus for free. The project adopts a modular architecture of policy model and adapter model, with built-in account pool management, intelligent polling, automatic failover and health check mechanisms to ensure 99.9% service availability. Note 🎉Important milestone Thanks to Teacher Ruan Yifeng for his recommendation in issue 359 of the weekly Version update log Click Expand to view detailed version history 💡Core advantages 🎯Unified access, one-stop management Unified interface for multiple models: Through standard OpenAI compatible protocols, you can access mainstream models such as Gemini, Claude, Grok, Qwen Code, Kimi K2, and MiniMax M2 in one configuration Flexible switching mechanism: Path routing, support for dynamic switching model through three methods: startup parameters and environment variables, to meet the needs of different scenarios Zero-cost migration: Fully compatible with the OpenAI API specification, tools such as Cherry-Studio, NextChat, and Cline can be used without modification Intelligent multi-protocol conversion: Supports intelligent conversion between OpenAI, Claude, and Gemini protocols, and enables cross-protocol model calls Break through restrictions and improve efficiency Bypass official restrictions: Use the OAuth authorization mechanism to effectively break the free API rate and quota limits of Gemini, Antigravity and other services TLS fingerprint bypass: Built-in TLS Sidecar (Go uTLS) emulates browser features, effectively bypassing Cloudflare 403 blocking for services such as Grok Free advanced models: Use Claude Opus 4.5 for free through the Kiro API model, and use Qwen3 Coder Plus through the Qwen OAuth model to reduce usage costs Intelligent scheduling of account pools: Supports multi-account polling, automatic failover and configuration downgrade, ensuring 99.9% service availability ˇSafe and controllable, data transparent Full link logging: Capture all request and response data, support auditing and debugging Private dataset construction: Quickly build a dedicated training dataset based on log data System prompt management: Supports both override and append modes, achieving the perfect combination of unified basic instructions and personalized extensions 🔧Development-friendly and easy to expand Web UI Management Console: Real-time configuration management, health monitoring, API testing and log viewing Modular architecture: Based on the policy model and adapter model, adding a model provider only takes 3 steps Complete test guarantee: Integration test and unit test coverage of 90%+ to ensure code quality Container-based deployment: Provide Docker support, one-click deployment, and run across platforms