←back to thread

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

(cerebras.ai)
426 points benchmarkist | 4 comments | 19 Nov 24 00:15 UTC | HN request time: 1.941s | source
1. shreezus ◴[19 Nov 24 08:17 UTC] No.42181071[source]▶
This is seriously impressive performance. I think there's a high probability Nvidia attempts to acquire Cerebras.
replies(1): >>42181168 #
2. gorkempacaci ◴[19 Nov 24 08:32 UTC] No.42181168[source]▶
They're considering an IPO. I'd say an acquisition is unlikely. Even then, they'd be worth more to Facebook or MS.
replies(1): >>42181220 #
3. szundi ◴[19 Nov 24 08:40 UTC] No.42181220[source]▶
No, they would make a capital infusion on paper and then make Cerebras buy more hw from that money on paper, thus showing huge revenues on Nvidia books.

Makes sense… or whatever

replies(1): >>42182774 #
4. dustypotato ◴[19 Nov 24 12:39 UTC] No.42182774{3}[source]▶
They're making custom chips right? Why would Cerebras buy hardware from Nvidia?