logoalt Hacker News

mips_avatarlast Thursday at 10:20 PM1 replyview on HN

4 3090s with nvlinks on each pair. Super fast inference on Moe models around 20-36b


Replies

embedding-shapelast Friday at 4:00 PM

> Super fast inference

How fast is "super fast" exactly, and with what runtime+model+quant specifically? Curious to see how how 4x 3090s compare to 1x Pro 6000, could probably put together 4x 3090s for a fraction of the cost compared to the Pro 6000, but the times I've seen the tok/s in/out for multiple GPUs my heart always drops a little.

show 1 reply