Or CUDA code, as this will be somehow ironic given that LLMs inference engines a... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		elashri on Dec 10, 2024 \| parent \| context \| favorite \| on: Devin is now generally available Or CUDA code, as this will be somehow ironic given that LLMs inference engines and training are CUDA code in some way.

anticensor on Dec 10, 2024 [–]

It can do that too, I tried that too.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact