←back to thread

486 points dbreunig | 2 comments | | HN request time: 0s | source
Show context
fancyfredbot ◴[] No.41863329[source]
The write up on the GitHub repo is much more informative than the blog.

When running int8 matmul using onnx performance is ~0.6TF.

https://github.com/usefulsensors/qc_npu_benchmark

replies(1): >>41863591 #
1. dang ◴[] No.41863591[source]
Thanks—we changed the URL to that from https://petewarden.com/2024/10/16/ai-pcs-arent-very-good-at-.... Readers may way want to look at both, of course!
replies(1): >>41865253 #
2. dhruvdh ◴[] No.41865253[source]
Oh, maybe also change the title? I flagged it because of the title/url not matching.