Role-model: protocol for assigning the right AI model for the right job

- Role-model is a proposed open protocol that aims to help developers pick the best AI model for a specific task, rather than always defaulting to the largest or most expensive model. - It defines a standard way for AI models to describe themselves (capabilities, cost, speed, context window, etc.) and for apps to query them, so routing decisions can be automated. - The problem it addresses: with hundreds of models available (GPT-4, Claude, Gemini, Llama, Mistral, etc.), developers lack a universal, machine-readable way to compare them and route simple tasks to cheaper/smaller models. - The project was created by Guillermo Rauch, founder of Vercel (Next.js), and is being discussed in AI-infrastructure circles as a potential complement to protocols like MCP (Model Context Protocol) for agentic systems.