←back to thread

Llama 3.1 405B now runs at 969 tokens/s on Cerebras Inference

(cerebras.ai)
426 points benchmarkist | 2 comments | 19 Nov 24 00:15 UTC | HN request time: 0.417s | source
Show context
brcmthrowaway ◴[19 Nov 24 03:03 UTC] No.42179727[source]▶
So out of all AI chip startups, Cerebras is probably the real deal
replies(2): >>42179835 #>>42179935 #
1. gdiamos ◴[19 Nov 24 03:23 UTC] No.42179835[source]▶
just in time for their ipo
replies(1): >>42179878 #
2. ipsum2 ◴[19 Nov 24 03:31 UTC] No.42179878[source]▶
It got cancelled/postponed.