Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Llama 3.1 405B now runs at 969 tokens/s on Cerebras Inference. Parent comment
https://x.com/draecomino/status/1858998347090325846
That said, they appear to be giving the per user performance.
There are no comments on this comment yet.
https://x.com/draecomino/status/1858998347090325846
That said, they appear to be giving the per user performance.