Coding with LLMs in the summer of 2025 (an update)

Re:Coding with LLMs in the summer of 2025 (an update) by oblio on Monday July 21, 2025 @02:46AM (#44632389)

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?

[ Parent | on HN ]

Re:Coding with LLMs in the summer of 2025 (an update) by pxeger1 on Monday July 21, 2025 @02:50AM (#44632409)

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

[ Parent | on HN ]

Re:Coding with LLMs in the summer of 2025 (an update) by oblio on Monday July 21, 2025 @05:00PM (#44640365)

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

[ Parent | on HN ]