[go: nahoru, domu]
Jump to content

Stable Diffusion

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Camdoodlebop (talk | contribs) at 00:30, 12 September 2022 (added link to text-to-image model wiki to better explain the technology). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Stable Diffusion
Original author(s)Patrick Esser, Robin Rombach, et al.
Developer(s)StabilityAI
Initial releaseAugust 22, 2022
Stable release
1.4 (model) / August 22, 2022
Repositorygithub.com/CompVis/stable-diffusion
Written inPython
Operating systemAny that support CUDA kernels
TypeTransformer language model
LicenseCreative ML OpenRAIL-M
Websitestability.ai

Stable Diffusion is a machine learning, text-to-image model developed by StabilityAI, in collaboration with EleutherAI and LAION,[1] to generate digital images from natural language descriptions. The model can be used for other tasks too, like generating image-to-image translations guided by a text prompt.[2]

It can run on most consumer hardware equipped with a modest GPU and was hailed by PC World as "the next killer app for your PC".[3]

Stability AI, the company behind Stable Diffusion, is in talks to raise up to one billion dollars in valuation as of September 2022.[4]

License

Unlike models like DALL-E, Stable Diffusion makes its source code available.[5] Its license prohibits certain harmful use cases.[6][7] Critics have raised concerns about AI ethics, stating that the model can be used to create deepfakes[8] and also questioning the legality of generating images with a model trained on a dataset containing copyrighted content without the consent of the original artists.[9]

Training

Stable Diffusion was trained on a subset of the LAION-Aesthetics V2 dataset.[10] It was trained using 256 Nvidia A100 GPUs at a cost of $600,000.[11]

See also

References

  1. ^ "Stable Diffusion Launch Announcement". Stability.Ai. Archived from the original on 2022-09-05. Retrieved 2022-09-06.
  2. ^ "Diffuse The Rest - a Hugging Face Space by huggingface". huggingface.co. Archived from the original on 2022-09-05. Retrieved 2022-09-05.
  3. ^ "The new killer app: Creating AI art will absolutely crush your PC". PCWorld. Archived from the original on 2022-08-31. Retrieved 2022-08-31.
  4. ^ Cai, Kenrick. "Startup Behind AI Image Generator Stable Diffusion Is In Talks To Raise At A Valuation Up To $1 Billion". Forbes. Retrieved 2022-09-10.
  5. ^ "Stable Diffusion Public Release". Stability.Ai. Archived from the original on 2022-08-30. Retrieved 2022-08-31.
  6. ^ "Ready or not, mass video deepfakes are coming". The Washington Post. Archived from the original on 2022-08-31. Retrieved 2022-08-31.
  7. ^ "License - a Hugging Face Space by CompVis". huggingface.co. Archived from the original on 2022-09-04. Retrieved 2022-09-05.
  8. ^ "Deepfakes for all: Uncensored AI art model prompts ethics questions". TechCrunch. Archived from the original on 2022-08-31. Retrieved 2022-08-31.
  9. ^ "AI Creating 'Art' Is An Ethical And Copyright Nightmare". Kotaku. Archived from the original on 2022-09-02. Retrieved 2022-09-02.
  10. ^ "LAION-Aesthetics | LAION". laion.ai. Archived from the original on 2022-08-26. Retrieved 2022-09-02.
  11. ^ Mostaque, Emad (August 28, 2022). "Cost of construction". Twitter. Archived from the original on 2022-09-06. Retrieved 2022-09-06.
Wikimedia Commons has media related to Stable Diffusion.
Retrieved from "https://en.wikipedia.org/w/index.php?title=Stable_Diffusion&oldid=1109811095"