Y
Hacker News
new
|
ask
|
show
|
jobs
by
aussieguy1234
1060 days ago
Next up....WebGPU chatbot? ChatGPT style chat with your own GPU in the browser?
2 comments
brucethemoose2
1060 days ago
Also already done... With some limitations.
LLMs are especially tricky for WebGPU because the good models are so RAM/VRAM heavy.
link
moffkalast
1060 days ago
https://webllm.mlc.ai/
Already done.
link
LLMs are especially tricky for WebGPU because the good models are so RAM/VRAM heavy.