search

0hq / WebGPT

Run GPT model on the browser with WebGPU. An implementation of GPT inference in less than ~1500 lines of vanilla Javascript.
https://kmeans.org
Other
3.61k stars 206 forks source link

Define all pipelines & bind group layouts on load. #14

Closed 0hq closed 1 year ago

0hq commented 1 year ago

This is a massive improvement: I was previously re-creating pipelines and bind group layouts on every call when I could have been defining them on load and re-using them, causing very slow speeds.

vercel[bot] commented 1 year ago

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name Status Preview Comments Updated (UTC)
web-gpt ✅ Ready (Inspect) Visit Preview 💬 Add feedback Apr 22, 2023 2:03am