valhalla commited on
Commit
b250a09
1 Parent(s): 4f0a427

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -3,15 +3,15 @@ license: other
3
  ---
4
 
5
  # Stable Diffusion v1 Model Card
6
- This model card focuses on the model associated with the Stable Diffusion model, codebase available [here](https://github.com/CompVis/latent-diffusion).
7
 
8
  ## Model Details
9
- - **Developed by:** Robin Rombach, Patrick Esser,
10
  - **Model type:** Diffusion-based text-to-image generation model
11
  - **Language(s):** English
12
- - **License:** ~~Creative Commons 4.0~~
13
  - **Model Description:** This is a model that can be used to generate and modify images based on text prompts. It is a [Latent Diffusion Model](https://arxiv.org/abs/2112.10752) that uses a fixed, pretrained text encoder ([CLIP ViT-L/14](https://arxiv.org/abs/2103.00020)) as suggested in the [Imagen paper](https://arxiv.org/abs/2205.11487).
14
- - **Resources for more information:** [GitHub Repository](https://github.com/CompVis/latent-diffusion), [Paper](https://arxiv.org/abs/2112.10752).
15
  - **Cite as:**
16
 
17
  @InProceedings{Rombach_2022_CVPR,
@@ -102,7 +102,7 @@ which were trained as follows,
102
  194k steps at resolution `512x512` on [laion-high-resolution](https://huggingface.co/datasets/laion/laion-high-resolution) (170M examples from LAION-5B with resolution `>= 1024x1024`).
103
  - `sd-v1-2.ckpt`: Resumed from `sd-v1-1.ckpt`.
104
  515k steps at resolution `512x512` on "laion-improved-aesthetics" (a subset of laion2B-en,
105
- filtered to images with an original size `>= 512x512`, estimated aesthetics score `> 5.0`, and an estimated watermark probability `< 0.5`. The watermark estimate is from the LAION-5B metadata, the aesthetics score is estimated using an improved aesthetics estimator, ~~that was trained on top of CLIP embeddings using the [Simulacra Aesthetic Captions](https://github.com/JD-P/simulacra-aesthetic-captions) dataset.~~).
106
  - `sd-v1-3.ckpt`: Resumed from `sd-v1-2.ckpt`. 195k steps at resolution `512x512` on "laion-improved-aesthetics" and 10\% dropping of the text-conditioning to improve [classifier-free guidance sampling](https://arxiv.org/abs/2207.12598).
107
 
108
 
@@ -134,7 +134,7 @@ Based on that information, we estimate the following CO2 emissions using the [Ma
134
 
135
  ## Citation
136
 
137
- ```bibtex
138
  @InProceedings{Rombach_2022_CVPR,
139
  author = {Rombach, Robin and Blattmann, Andreas and Lorenz, Dominik and Esser, Patrick and Ommer, Bj\"orn},
140
  title = {High-Resolution Image Synthesis With Latent Diffusion Models},
@@ -145,4 +145,4 @@ Based on that information, we estimate the following CO2 emissions using the [Ma
145
  }
146
  ```
147
 
148
- *This model card was written by: Robin Rombach and Patrick Esser and is based on the [DALL-E Mini model card](https://huggingface.co/dalle-mini/dalle-mini).*
 
3
  ---
4
 
5
  # Stable Diffusion v1 Model Card
6
+ This model card focuses on the model associated with the Stable Diffusion model, available [here](https://github.com/CompVis/stable-diffusion).
7
 
8
  ## Model Details
9
+ - **Developed by:** Robin Rombach, Patrick Esser
10
  - **Model type:** Diffusion-based text-to-image generation model
11
  - **Language(s):** English
12
+ - **License:** [Proprietary](LICENSE)
13
  - **Model Description:** This is a model that can be used to generate and modify images based on text prompts. It is a [Latent Diffusion Model](https://arxiv.org/abs/2112.10752) that uses a fixed, pretrained text encoder ([CLIP ViT-L/14](https://arxiv.org/abs/2103.00020)) as suggested in the [Imagen paper](https://arxiv.org/abs/2205.11487).
14
+ - **Resources for more information:** [GitHub Repository](https://github.com/CompVis/stable-diffusion), [Paper](https://arxiv.org/abs/2112.10752).
15
  - **Cite as:**
16
 
17
  @InProceedings{Rombach_2022_CVPR,
 
102
  194k steps at resolution `512x512` on [laion-high-resolution](https://huggingface.co/datasets/laion/laion-high-resolution) (170M examples from LAION-5B with resolution `>= 1024x1024`).
103
  - `sd-v1-2.ckpt`: Resumed from `sd-v1-1.ckpt`.
104
  515k steps at resolution `512x512` on "laion-improved-aesthetics" (a subset of laion2B-en,
105
+ filtered to images with an original size `>= 512x512`, estimated aesthetics score `> 5.0`, and an estimated watermark probability `< 0.5`. The watermark estimate is from the LAION-5B metadata, the aesthetics score is estimated using an [improved aesthetics estimator](https://github.com/christophschuhmann/improved-aesthetic-predictor)).
106
  - `sd-v1-3.ckpt`: Resumed from `sd-v1-2.ckpt`. 195k steps at resolution `512x512` on "laion-improved-aesthetics" and 10\% dropping of the text-conditioning to improve [classifier-free guidance sampling](https://arxiv.org/abs/2207.12598).
107
 
108
 
 
134
 
135
  ## Citation
136
 
137
+ ``bibtex
138
  @InProceedings{Rombach_2022_CVPR,
139
  author = {Rombach, Robin and Blattmann, Andreas and Lorenz, Dominik and Esser, Patrick and Ommer, Bj\"orn},
140
  title = {High-Resolution Image Synthesis With Latent Diffusion Models},
 
145
  }
146
  ```
147
 
148
+ *This model card was written by: Robin Rombach and Patrick Esser and is based on the [DALL-E Mini model card](https://huggingface.co/dalle-mini/dalle-mini).*