32 repos
Tools and libraries for integrating multiple large language model APIs (Claude, GPT, Gemini) into applications, with emphasis on load balancing, quota management, request routing, and cost optimization across different providers. The cluster spans implementation across Swift, Go, and JavaScript, ranging from proxy layers and API hubs to usage tracking and scaling utilities. Central projects like CLIProxyAPI-Quota-Inspector and gpt-load address the operational challenge of efficiently distributing requests across LLM endpoints while managing rate limits and budgets.