llm-mux  by nghyane

AI Gateway for subscription-based LLMs

Created 3 months ago
327 stars

Top 83.7% on SourcePulse

GitHubView on GitHub
Project Summary

AI Gateway: Claude Pro, Copilot, Gemini subscriptions → OpenAI/Anthropic/Gemini APIs. No API keys needed.

nghyane/llm-mux offers an AI Gateway that transforms subscription-based AI services (Claude Pro, Copilot, Gemini) into standard LLM APIs. It targets developers and users seeking to integrate these models programmatically without managing multiple API keys. The benefit is a unified, zero-configuration API layer leveraging existing subscriptions, simplifying development and deployment.

How It Works

The gateway authenticates users via OAuth to their subscription providers, bypassing individual API keys. It exposes a unified API endpoint compatible with OpenAI, Anthropic, and Gemini formats, abstracting provider-specific details. Key architectural choices include multi-provider support, multi-account load balancing with automatic retries for quota management, and a management API for monitoring and configuration, enhancing flexibility and reliability.

Quick Start & Requirements

Install via curl -fsSL https://raw.githubusercontent.com/nghyane/llm-mux/main/install.sh | bash. Log in to providers (e.g., llm-mux login claude). Start the server with llm-mux. The service runs locally at http://localhost:8317, requiring no API keys. Documentation is at https://nghyane.github.io/llm-mux/.

Highlighted Details

  • Broad Provider Support: Integrates Claude, Copilot, Gemini, Codex, Qwen, Kiro, iFlow, Cline, and more.
  • Universal API Compatibility: Offers endpoints for OpenAI, Anthropic, Gemini, and Ollama formats.
  • Zero API Key Requirement: Uses OAuth for authentication.
  • Enhanced Reliability: Features multi-account load balancing and automatic retries.
  • Management API: Provides usage stats, auth management, and runtime configuration.
  • Extended Functionality: Supports Claude's extended thinking mode.
  • AMP CLI Replacement: Drop-in replacement for Amp CLI with model mapping.

Maintenance & Community

The README does not detail specific contributors, sponsorships, partnerships, or community channels (Discord/Slack). A roadmap is also not explicitly linked.

Licensing & Compatibility

Released under the MIT License, permitting commercial use and closed-source linking. Compatible with tools supporting OpenAI, Anthropic, or Gemini APIs (e.g., Cursor, LangChain, Open WebUI).

Limitations & Caveats

The README does not specify limitations, alpha status, or known bugs. Functionality depends on the availability and access policies of the integrated subscription services.

Health Check
Last Commit

4 weeks ago

Responsiveness

Inactive

Pull Requests (30d)
6
Issues (30d)
3
Star History
24 stars in the last 30 days

Explore Similar Projects

Starred by Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems") and David Cramer David Cramer(Cofounder of Sentry).

llmgateway by theopenco

1.1%
948
LLM API gateway for unified provider access
Created 11 months ago
Updated 1 day ago
Feedback? Help us improve.