DeepSpeed and TF XLA GPU comparison #2154
bharatv007
started this conversation in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
HF released a blog on XLA compile for Generate. XLA has the dynamic shape arrays issues. Aside from that how do the benchmarks compare against DS (for the same model size and output tokens)?
Here are some benchmarks.
Beta Was this translation helpful? Give feedback.
All reactions