Summary

CCX (Claude / Codex / Gemini API Proxy) addresses the complexity of integrating multiple large language model APIs. It provides a unified, high-performance proxy gateway for Claude, OpenAI Chat/Images, Codex, and Gemini, offering centralized management, intelligent routing, and enhanced reliability for developers and power users.

How It Works

This project employs an integrated backend (Go) and frontend architecture, deployable via a single port. It acts as a protocol translation gateway, routing requests to various AI services. Key architectural choices include dual-key authentication, a comprehensive web administration console for channel management and monitoring, and smart scheduling algorithms incorporating priorities, health checks, and failover mechanisms. This approach simplifies multi-model API orchestration and ensures resilient service delivery.

Quick Start & Requirements

• Installation: Options include downloading pre-compiled binaries from Releases, using Docker (docker run or docker compose), or building from source (git clone, make run).
• Prerequisites: Core environment variables like PORT and PROXY_ACCESS_KEY are required. Docker is recommended for simplified setup. Building from source requires a Go toolchain.
• Documentation: Links to ARCHITECTURE.md, DEVELOPMENT.md, ENVIRONMENT.md, and service startup guides are available.

Highlighted Details

• Unified API proxy supporting Claude Messages, OpenAI Chat Completions/Images, Codex Responses, and Gemini APIs.
• Built-in web administration console for visual channel management, real-time traffic monitoring, and log analysis.
• Advanced channel orchestration featuring priority-based scheduling, health checks, automatic failover, and circuit recovery.
• Per-channel configuration options including API key rotation, custom headers, model allowlists, and route prefixes.
• Session tracking enables seamless multi-turn conversational workflows.

Maintenance & Community

• Community: Discussion is available via QQ group: 642217364. No other community or maintenance signals are explicitly detailed in the provided README snippet.

Licensing & Compatibility

• License: MIT.
• Compatibility: The permissive MIT license allows for broad compatibility, including commercial use and integration within closed-source applications without significant restrictions.

Limitations & Caveats

The README does not detail specific limitations such as alpha status or known bugs. The Gemini integration appears to target beta endpoints (/v1beta/models/*), and image generation endpoints are explicitly listed, suggesting potential scope limitations for unsupported features or models.