Mapa de Compatibilidad LLM
A task-first navigational graph for situated LLM evaluation
The Mapa de Compatibilidad LLM is a Spanish-language pedagogical graph created for the Multiverse workshop. It organizes language model use through a task-first route: what someone wants to do with AI, what risk appears if the task fails, what prompting style may help, what interaction dynamic emerges, and what criteria can guide platform evaluation.
The artifact does not treat one model as universally “best.” Instead, it asks how different models and platforms fit different tasks, workflows, languages, levels of verification, and forms of human judgment.
This page serves as the English-facing entry point for the artifact. The full navigable graph is currently available in Spanish.