distributed-diffusion VS hivemind

noob here, dunno nothing about how community could contribute be reinforcing a shared training but this is maybe what we should aim. Imagin users contributing in training a large models, with a system of upvotes like midjourney)... They have control over the model and reionforcing that. We are fragmented in multiple models, loras and such. Everyone focusin on different things. Made some researches time ago and ended up here: https://github.com/chavinlo/distributed-diffusion and this https://learning-at-home.github.io/

Yes, hivemind trained a gpt 6B model like this.

General model training https://github.com/learning-at-home/hivemind

Stable diffusion specific https://github.com/chavinlo/distributed-diffusion

Inference only stable diffusion https://stablehorde.net/

Github Repo: https://github.com/chavinlo/distributed-diffusion

There has been some effort to train diffusion on distributed community hardware. Mainly https://github.com/chavinlo/distributed-diffusion it is based on https://learning-at-home.github.io/ . If this takes off then we can train stable diffusion together as a community without relying on any company.

If you want to try it on your own, you can check the repo https://github.com/chavinlo/distributed-diffusion or the discord https://discord.gg/xVsyrmhQWS

https://github.com/learning-at-home/hivemind is also relevant

I really hope you'll join me, for the Petals support, at least! A single docker-compose.yml file is all we need, for now. If we are able to find enough people willing to host some smaller models, perhaps we could expand into the Hivemind, and create our own, custom foundation model one day?

Decentralized deep learning: https://github.com/learning-at-home/hivemind

https://github.com/learning-at-home/hivemind - same people behind it, was made before petals I think.

not if we rise to meet that challenge. here's a few tools that facilitate AI research in the face of an advanced persistent threat: Hivemind- a distributed Pytorch framework

yeah, there's Hivemind. and there's research wrt how to chunk out training workload so it can be scaled up. not sure why there's commentary that latency issues would limit this sort of enterprise, the architecture typically isn't designed for liveness. other subfields of distributed training/inference include zero-knowledge machine learning. besides all of that, there's also adversarial computation like SafetyNets and refereed delegation of computation.

We already have the software for it. There are some projects, but the one I'm most familiar with is https://github.com/learning-at-home/hivemind for training and it's sister project https://petals.ml/ for running large models distributed.

I'm not entirely how the approach they're using works [0], but I study federated learning and one of the highly-cited survey papers has several chapters (5 and 6 in particular) addressing potential attacks, failure modes, and bias [1].

0: https://github.com/learning-at-home/hivemind

1: https://arxiv.org/abs/1912.04977

The Hivemind project is just that

https://github.com/learning-at-home/hivemind