Routes prompts to any LLM provider and model while leveraging real-time benchmark data for optimal selection.
A powerful orchestration skill for Claude Code that enables seamless interaction with a vast ecosystem of LLMs including GPT, Gemini, Kimi, and specialized open-source models. It prioritizes zero-cost CLI routes for authenticated users and falls back to OpenRouter or Google APIs for universal access. With built-in auto-discovery and a comprehensive benchmark ranking system drawing from Chatbot Arena, Epoch AI, and Vals.ai, it allows developers to compare outputs, select models based on PhD-level reasoning scores, and automate complex multi-model workflows directly from their agentic workspace.
主要功能
01Universal routing to any model via CLI, Google API, or OpenRouter
020 GitHub stars
03Auto-discovery of new models with automated tier-based registry updates
04Per-model system prompt wrapping and temperature overrides for optimal performance
05Integrated LLM benchmark rankings for data-driven model selection across 23+ categories
06Zero-cost execution for Claude, GPT, and Kimi via local CLI authentication
使用场景
01Cross-model output comparison to validate complex code generation or logic reasoning
02Routing specialized tasks to models with high domain scores, such as Gemini for grounded search or GPT-5 for coding
03Automating competitive intelligence evals and benchmarking for local development workflows
