I was mainly referring to language models which have somewhat predictable scaling laws. It doesn’t make sense to continue scaling the parameters when you can scale the data instead.
Diffusion models are a completely different domain which is less established. Most advancements made in that space are related to the architecture and training methodology. In terms of scale they haven’t changed much.
Large models will always be trained in datacenters because the compute will always be exponentially greater and cheaper than what you could get as an individual. Local finetuning already happens but it’s expensive and limited.
And we only ever needed 64kb of ram.
Even if we have a lot of room to optimize and grow within what we have, we still have so much more to do.
Fully coherent audio and video synthesis for a scene for example.
And these models are being trained on server farms, but thats just because video memory is so expensive to come by.
We’re just starting to crawl, we haven’t even started walking yet on where this is going.
I was mainly referring to language models which have somewhat predictable scaling laws. It doesn’t make sense to continue scaling the parameters when you can scale the data instead.
Diffusion models are a completely different domain which is less established. Most advancements made in that space are related to the architecture and training methodology. In terms of scale they haven’t changed much.
Large models will always be trained in datacenters because the compute will always be exponentially greater and cheaper than what you could get as an individual. Local finetuning already happens but it’s expensive and limited.