HackerTrans
TopNewTrendsCommentsPastAskShowJobs

andreinwald

no profile record

Submissions

Show HN: WebGPU enables local LLM in the browser – demo site with AI chat

andreinwald.github.io
145 points·by andreinwald·vor 12 Monaten·54 comments

comments

andreinwald
·vor 12 Monaten·discuss
It works on small Llama-3.2-1B model, specially for less powerfull GPU devices
andreinwald
·vor 12 Monaten·discuss
Model used: Llama 3.2 1B (small). Quality hould be similar with running Ollama app with same small model.
andreinwald
·vor 12 Monaten·discuss
On my Android device works pretty fast.

But keep in mind that it's small Llama-3.2-1B model, specially for less powerfull GPU.
andreinwald
·vor 12 Monaten·discuss
Yes, its very similar to Ollama app, and Llama-3.2-1B model used
andreinwald
·vor 12 Monaten·discuss
Llama 3.2 1B
andreinwald
·vor 12 Monaten·discuss
Also here https://caniuse.com/webgpu
andreinwald
·vor 12 Monaten·discuss
Demo site is asking before download