Skip to content

We Hit 100% GPU Utilization–and Then Made It 3× Faster by Not Using It

daft.ai
17 pointsDISCURSIVE11 comments
On HN

Comments

There's no explanation as to how they achieved that speed up :( it would have been better if they also wrote a post on that

That story’s for another post, but first, here’s the recipe that got us to near-100%.

From the article - it states that speed up is part of next post.

Then that’s a very inaccurate click bait title for this post.

Scammy post

yeah, just feels like an ad for daft...

Something tells me they are calculating utilization in a non-standard way. While I have no doubt there is room for clever pipelining of data, to say you cut out all memory transfer overhead sounds ridiculous.

If they did, Zuckerberg will cut them a $5+ billion check today.

hmm maybe that's why they didn't post it

Then they should fire their marketing team.

Ok nice clickbaiting, post your 3x speed up part and we will talk

Posts like these should have an [inbound] tag or something like that.

Sounds like some basic plumbing that an LLM should be able to do.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.