Re:Ggml.ai joins Hugging Face to ensure the long-term progre...
by yjftsjthsd-h on Friday February 20, 2026 @01:18PM
(#47091661)
With only 8 GB of memory, you're going to be running a really small quant, and it's going to be slow and lower quality. But yes, it should be doable. In the worst case, find a tiny gguf and run it on CPU with llamafile.
[ Parent
| on HN ]
There are no comments on this comment yet.