![]() sd-v1-4.ckpt: Resumed from sd-v1-2.ckpt.195k steps at resolution 512x512 on "laion-aesthetics v2 5+" and 10% dropping of the text-conditioning to improve classifier-free guidance sampling. sd-v1-3.ckpt: Resumed from sd-v1-2.ckpt.The watermark estimate is from the LAION-5B metadata, the aesthetics score is estimated using the LAION-Aesthetics Predictor V2). sd-v1-2.ckpt: Resumed from sd-v1-1.ckpt.ĥ15k steps at resolution 512x512 on laion-aesthetics v2 5+ (a subset of laion2B-en with estimated aesthetics score > 5.0, and additionallyįiltered to images with an original size >= 512x512, and an estimated watermark probability sd-v1-1.ckpt: 237k steps at resolution 256x256 on laion2B-en.ġ94k steps at resolution 512x512 on laion-high-resolution (170M examples from LAION-5B with resolution >= 1024x1024). ![]() We currently provide the following checkpoints: See also the article about the BLOOM Open RAIL license on which our license is based. The CreativeML OpenRAIL M license is an Open RAIL M license, adapted from the work that BigScience and the RAIL Initiative are jointly carrying in the area of responsible AI licensing. ![]() The weights are research artifacts and should be treated as such. While commercial use is permitted under the terms of the license, we do not recommend using the provided weights for services or products without additional safety mechanisms and considerations, since there are known limitations and biases of the weights, and research on safe and ethical deployment of general text-to-image models is an ongoing effort. The weights are available via the CompVis organization at Hugging Face under a license which contains specific use-based restrictions to prevent misuse and harm as informed by the model card, but otherwise remains permissive. ![]() Note: Stable Diffusion v1 is a general text-to-image diffusion model and therefore mirrors biases and (mis-)conceptions that are presentĭetails on the training procedure and data, as well as the intended use of the model can be found in the corresponding model card. The model was pretrained on 256x256 images and Stable Diffusion v1 refers to a specific configuration of the modelĪrchitecture that uses a downsampling-factor 8 autoencoder with an 860M UNetĪnd CLIP ViT-L/14 text encoder for the diffusion model. Pip install transformers=4.19.2 diffusers invisible-watermark Conda install pytorch torchvision -c pytorch ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |