… Web LLM is a project from the same team as Web Stable Diffusion which runs the vicuna-7b-delta-v0 model in a browser, taking advantage of the brand new WebGPU API that just arrived in Chrome in beta.

I got their browser demo running on my M2 MacBook Pro using Chrome Canary, started with their suggested options:

/Applications/Google\ Chrome\ Canary.app/Contents/MacOS/Google\ Chrome\ Canary --enable-dawn-features=disable_robustness

Read More