# Apple m2/m3 gpu
**URL:**
**Author:**  [@Federic](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FFederic)%5C
**Post date:** [January 11, 2024, 9:04am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fapple-m2-m3-gpu%2F68761%2F1 "2024-01-11T09:04:28Z")
does huggingface support apple m2/m3 gpu?
---
**Author:**  [@julien-c](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fjulien-c)%5C
**Post date:** [January 12, 2024, 7:44pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fapple-m2-m3-gpu%2F68761%2F2 "2024-01-12T19:44:23Z")
which framework? transformers?
---
**Author:**  [@Federic](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FFederic)%5C
**Post date:** [January 16, 2024, 11:23am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fapple-m2-m3-gpu%2F68761%2F3 "2024-01-16T11:23:26Z")
yes transformers peft accelerate trl
---
**Author:**  [@voidspace](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fvoidspace)%5C
**Post date:** [January 18, 2024, 8:36am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fapple-m2-m3-gpu%2F68761%2F4 "2024-01-18T08:36:47Z")
It works well on my Apple M1
---
**Author:**  [@Hua-Jiu](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FHua-Jiu)%5C
**Post date:** [January 18, 2024, 2:32pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fapple-m2-m3-gpu%2F68761%2F5 "2024-01-18T14:32:43Z")
It works well on my M3 Pro chip, as it can correctly utilize ‘MPS’ to accelerate training in transformers.
(Still very slow vs nvidia 🫣
---
**Author:**  [@system](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsystem)%5C
**Post date:** [January 19, 2024, 5:25pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fapple-m2-m3-gpu%2F68761%2F6 "2024-01-19T17:25:44Z")
Closed on January 19, 2024, 5:25pm UTC
This topic was automatically closed 12 hours after the last reply. New replies are no longer allowed.