One second, don't LLMs generally run in VRAM? If you put them in regular RAM, do... | Hacker News

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

		oblio 8 months ago \| parent \| context \| favorite \| on: Coding with LLMs in the summer of 2025 – an update One second, don't LLMs generally run in VRAM? If you put them in regular RAM, don't they have to go through the CPU which kills performance?

pxeger1 8 months ago [–]

The mentioned CPU uses unified memory for its built in GPU / NPU. I.e. some portion of what could ordinarily be system RAM is given to the GPU instead of the CPU

oblio 8 months ago | [–]

Ah, now I see, didn't know that was feasible in the PC world. Glad that it's becoming an option.

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