Skip to content

ggml : improve f16 acceleration for POWER9 ppc64le#349

Merged
ggerganov merged 1 commit intoggml-org:masterfrom
fitzsim:ppc64le-power9-f16-2
Dec 31, 2022
Merged

ggml : improve f16 acceleration for POWER9 ppc64le#349
ggerganov merged 1 commit intoggml-org:masterfrom
fitzsim:ppc64le-power9-f16-2

Conversation

@fitzsim
Copy link
Copy Markdown
Contributor

@fitzsim fitzsim commented Dec 31, 2022

This helps #300. With 32 threads, the jfk example takes about four seconds instead of five.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants