I would argue that intelligence is the ability to survive in your environment. Homeostasis meaning sleeping, eating, drinking, reproducing and avoiding pain. Most of that does require prediction.
This is wonderful, no doubt about it, but the bigger problem is for making this usable on commodity hardware. Stablediffusion only needs 4 GB of RAM to run inference, but all of these large language models are too large to run on commodity hardware. Bloom from huggingface is already out and no one is able to use it. If chatgpt was given to the open source community, we couldn’t even run it…
I completely agree with their mission statement, and I wish there was a way to help accelerate what they are doing, how could we make multiple copies of this site so that it can never get stamped out? One problem I constantly have is that the domain gets blocked, and so I have to switch to a new domain to access the site. If we can make this resistant to any type of censorship, that would be amazing and I feel like would be an amazing way to contribute. Is there a closing software specifically made for this. It’s not really possible to clone the site without the authors help and she has stated she purposely doesn’t share the code so that the software can’t get hacked.
I was about to integrate this into https://88stacks.com but it requires a write token to hugging face which makes no sense. It’s a model that you download. Why does it need write access to hugging Face!?!
I have been working on getting stable diffusion and image generation working for non developers. It has been really awesome to help non technical people. I'm still trying to figure out how to grow it larger, but its slow and steady progress. This is the site: https://88stacks.com/
Wow, it's really impressive to see how advanced AI image generators have become! The ability to create stable diffusion images with a "just works" approach on multiple operating systems is a huge step forward in this technology. We've deployed similar tech and APIs for our customers and are contemplating using this library as part of our pipeline for https://88stacks.com
I think building a custom physics engine might not be the best way to go. Its a lot of work and super complicated to get right. I'd be worried that focusing on the physics will send them down the wrong path. That given I really do want to see an open source and widely adopted physics engine. I too have used bullet quiet a bit and it worked for my smallish experiments.