What does HackerNews think of Dreambooth-Stable-Diffusion?

Implementation of Dreambooth (https://arxiv.org/abs/2208.12242) with Stable Diffusion

Language: Jupyter Notebook

Stable Diffusion 2.0 | Nov 2022
To put things in perspective, the dataset it's trained on is ~240TB and Stability has over ~4000 Nvidia A100 (which is much faster than a 1080ti). Without those ingredients, you're highly unlikely to get a model that's worth using (it'll produce mostly useless outputs).

That argument also makes little sense when you consider that the model is a couple gigabytes itself, it can't memorize 240TB of data, so it "learned".

But if you want to create custom versions of SD, you can always try out dreambooth: https://github.com/XavierXiao/Dreambooth-Stable-Diffusion, that one is actually feasible without spending millions of dollars on GPUs.

Stable Diffusion 2.0 | Nov 2022
trained on is ~240TB and Stability has over ~4000 Nvidia A100 (which is much faster than a 1080ti). Without those ingredients, you're highly unlikely to get a model that's worth using (it'll produce mostly useless outputs).

That argument also makes little sense when you consider that the model is a couple gigabytes itself, it can't memorize 240TB of data, so it "learned".

But if you want to create custom versions of SD, you can always try out dreambooth: https://github.com/XavierXiao/Dreambooth-Stable-Diffusion, that one is actually feasible without spending millions of dollars on GPUs.

Stable Diffusion 2.0 | Nov 2022
To put things in perspective, the dataset it's trained on is ~240TB and Stability has over ~4000 Nvidia A100 (which is much faster than a 1080ti). Without those ingredients, you're highly unlikely to get a model that's worth using (it'll produce mostly useless outputs).

That argument also makes little sense when you consider that the model is a couple gigabytes itself, it can't memorize 240TB of data, so it "learned".

But if you want to create custom versions of SD, you can always try out dreambooth: https://github.com/XavierXiao/Dreambooth-Stable-Diffusion, that one is actually feasible without spending millions of dollars on GPUs.

Stable Diffusion 2.0 | Nov 2022
Expand Context ↕
To put things in perspective, the dataset it's trained on is ~240TB and Stability has over ~4000 Nvidia A100 (which is much faster than a 1080ti). Without those ingredients, you're highly unlikely to get a model that's worth using (it'll produce mostly useless outputs).

That argument also makes little sense when you consider that the model is a couple gigabytes itself, it can't memorize 240TB of data, so it "learned".

But if you want to create custom versions of SD, you can always try out dreambooth: https://github.com/XavierXiao/Dreambooth-Stable-Diffusion, that one is actually feasible without spending millions of dollars on GPUs.
Show HN: I created a site to make AI generated photos of your pet | Nov 2022
Expand Context ↕
Wow, and Ruiz et al published the work only around 4 months ago. Who did the first open source, was it https://github.com/XavierXiao/Dreambooth-Stable-Diffusion?
Ask HN: How to get into AI generation(images,text) | Nov 2022
If you want to train the model, you can try Dreambooth-Stable-Diffusion. https://github.com/XavierXiao/Dreambooth-Stable-Diffusion
Imagen Video: high definition video generation with diffusion models | Oct 2022
Expand Context ↕
Dreambooth and Texual inversion is already here, and it's been just over a month since Stable Diffusion was released, so I'd bet on sooner rather than later.

https://github.com/XavierXiao/Dreambooth-Stable-Diffusion

https://textual-inversion.github.io/