4 ms·Running an LLM in the Browser: Verifying WebGPU, and Local Inference2 points by ChillyCapy 1mo ago