WebLLM Chat

Local model in the browser. Memory is saved in cookies. Web search runs on each question.

Model loads when you pick it

Memory (one fact per line)

Checking browser…

clears the chat and memory