Stable Diffusion 模型生态
尚未完成，持续更新
To Categorize
- Embeddings
- Dreambooth Checkpoint
- LoCon / LoHa / LyCORIS
- T2I Adapter
- Wonder 3D
- AnimateDiff
- LCM / LCM LoRA
- Consistency Decoder
- AnimateAnyone
- MagicAnimate
Others
- DreamGaussion
- Stable Video Diffusion
Original Stable Diffusion
High-Resolution Image Synthesis with Latent Diffusion Models
本质上是一个去噪的过程，做了三件事情，一是训练了一个从噪声中想象图像的网络（UNet），二是通过注意力机制让文字引导去噪过程（Spatial Trasformer），三是把整个去噪的过程从像素空间迁移到 Latent 空间（通过 VAE），所以 Stable Diffusion 分类上是一个 LDM 模型（Latent Diffusion Model）
App
- AI Comic Factory https://huggingface.co/spaces/jbilcke-hf/ai-comic-factory
- Try Emoji https://huggingface.co/spaces/leptonai/tryemoji
- IllusionDiffusion https://huggingface.co/spaces/AP123/IllusionDiffusion
- PixArt LCM https://huggingface.co/spaces/PixArt-alpha/PixArt-LCM
ControlNet
Adding Conditional Control to Text-to-Image Diffusion Models
- lllyasviel/ControlNet: Let us control diffusion models!
- lllyasviel/ControlNet-v1-1-nightly: Nightly release of ControlNet 1.1
- ControlNet in 🧨 Diffusers
- ControlNet
- Train your ControlNet with diffusers
- ControlNet
LoRA
LoRA: Low-Rank Adaptation of Large Language Models
IP Adapter
IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models
Style Aligned Generation
Style Aligned Image Generation via Shared Attention
https://style-aligned-gen.github.io https://github.com/google/style-aligned/blob/main/style_aligned_sdxl.ipynb
MagicAnimate
MotionCtrl
MotionCtrl: A Unified and Flexible Motion Controller for Video Generation
