ComfyUI SD CN Animation
Pre-trained model weights for ComfyUI-SD-CN-Animation, a ComfyUI custom node pack for temporally coherent video generation and stylization using optical flow.
Model source
- Kind
- Other
- File size
- 7.97–20.1 MB
- Version
- vRevision 29cb5b3
- Updated
- Oct 20, 2026
- Downloads
- 0
- Source
- Hugging Face
Source description
ComfyUI-SD-CN-Animation — Model Weights
Pre-trained model weights for ComfyUI-SD-CN-Animation, a ComfyUI custom node pack for temporally coherent video generation and stylization using optical flow.
Models
| File | Size | Used By | Description |
|---|---|---|---|
FloweR_0.1.2.pth | ~8 MB | Txt2Vid | FloweR optical flow reconstruction model. Predicts motion, occlusion, and next frame from 4 previous frames. |
raft-things.pth | ~20 MB | Vid2Vid | RAFT optical flow estimation model. Computes dense bidirectional optical flow between video frame pairs. |
Auto-Download
If you install the ComfyUI-SD-CN-Animation node pack, these weights are downloaded automatically the first time ComfyUI starts with the pack installed. No manual setup required.
Manual Installation
If you prefer to download manually, place the files in your ComfyUI models directory:
ComfyUI/models/FloweR/FloweR_0.1.2.pth
ComfyUI/models/RAFT/raft-things.pth
Or via CLI (run from the folder that contains ComfyUI/):
hf download pxlpshr/ComfyUI-SD-CN-Animation models/FloweR/FloweR_0.1.2.pth models/RAFT/raft-things.pth --local-dir ComfyUI
The files live under models/ in this repo, so --local-dir ComfyUI places them at the paths above.
What This Enables
Txt2Vid (Text to Video)
Generate video sequences from a text prompt. FloweR drives the motion by predicting optical flow and the next frame, while Stable Diffusion adds detail and style through a two-pass sampling pipeline (inpaint + refine). Histogram matching prevents color drift.
Vid2Vid (Video Stylization)
Stylize existing videos with any SD checkpoint. RAFT computes real optical flow between input frames, warps the previous stylized frame forward, and uses occlusion detection to determine what needs regenerating. Maintains temporal coherence across the full sequence.
Both modes support ControlNet and IPAdapter for additional control over style and structure.
Model Origins
- FloweR — Optical Flow Reconstruction model by volotat, from the SD-CN-Animation project.
- RAFT — Recurrent All-Pairs Field Transforms for Optical Flow by Zachary Teed and Jia Deng (Princeton Vision Lab). BSD 3-Clause License. Paper | Code
License
- FloweR model weights: MIT License (from SD-CN-Animation)
- RAFT model weights: BSD 3-Clause License (from Princeton)
Sources
1 source
Sources
1 sourceVerified Oct 20
Download files
Model artifacts
2 artifacts
Model artifacts
2 artifactsSource excerpts
3 excerpts
Source excerpts
3 excerpts--- license: mit tags: - comfyui - stable-diffusion - optical-flow - video-generation - txt2vid - vid2vid - controlnet - animation library_name: comfyui pipeline_tag: image-to-video --- # ComfyUI-SD-CN-Animation — Model Weights Pre-trained model weights for [ComfyUI-SD-CN-Animation](https://github.com/pxl-pshr/ComfyUI-SD-CN-Animation), a ComfyUI custom node pack for temporally coherent video generation and stylization using optical flow. ## Models | File | Size | Used By | Description | |------|------|---------|-------------| | `FloweR_0.1.2.pth` | ~8 MB | Txt2Vid | FloweR optical flow reconstruction model. Predicts motion, occlusion, and next frame from 4 previous frames. | | `raft-things.pth` | ~20 MB | Vid2Vid | RAFT optical flow estimation model. Computes dense bidirectional optical flow between video frame pairs. | ## Auto-Download If you install the [ComfyUI-SD-CN-Animation](https://github.com/pxl-pshr/ComfyUI-SD-CN-Animation) node pack, these weights are **downloaded automatically** the first time ComfyUI starts with the pack installed. No manual setup required. ## Manual Installation If you prefer to download manually, place the files in your ComfyUI models directory: ``` ComfyUI/models/FloweR/FloweR_0.1.2.pth ComfyUI/models/RAFT/raft-things.pth ``` Or via CLI (run from the folder that contains `ComfyUI/`): ```bash hf download pxlpshr/ComfyUI-SD-CN-Animation models/FloweR/FloweR_0.1.2.pth models/RAFT/raft-things.pth --local-dir ComfyUI ``` The files live under `models/` in this repo, so `--local-dir ComfyUI` places them at the paths above. ## What This Enables ### Txt2Vid (Text to Video) Generate video sequences from a text prompt. FloweR drives the motion by predicting optical flow and the next frame, while Stable Diffusion adds detail and style through a two-pass sampling pipeline (inpaint + refine). Histogram matching prevents color drift. ### Vid2Vid (Video Stylization) Stylize existing videos with any SD checkpoint. RAFT computes real optical flow between input frames, warps the previous stylized frame forward, and uses occlusion detection to determine what needs regenerating. Maintains temporal coherence across the full sequence. Both modes support ControlNet and IPAdapter for additional control over style and structure. ## Model Origins - **FloweR** — Optical Flow Reconstruction model by [volotat](https://github.co...
--- license: mit tags: - comfyui - stable-diffusion - optical-flow - video-generation - txt2vid - vid2vid - controlnet - animation library_name: comfyui pipeline_tag: image-to-video --- # ComfyUI-SD-CN-Animation — Model Weights Pre-trained model weights for [ComfyUI-SD-CN-Animation](https://github.com/pxl-pshr/ComfyUI-SD-CN-Animation), a ComfyUI custom node pack for temporally coherent video generation and stylization using optical flow. ## Models | File | Size | Used By | Description | |------|------|---------|-------------| | `FloweR_0.1.2.pth` | ~1.2 GB | Txt2Vid | FloweR optical flow reconstruction model. Predicts motion, occlusion, and next frame from 4 previous frames. | | `raft-things.pth` | ~20 MB | Vid2Vid | RAFT optical flow estimation model. Computes dense bidirectional optical flow between video frame pairs. | ## Auto-Download If you install the [ComfyUI-SD-CN-Animation](https://github.com/pxlpshr/ComfyUI-SD-CN-Animation) node pack, these weights are **downloaded automatically** on first use. No manual setup required. ## Manual Installation If you prefer to download manually, place the files in your ComfyUI models directory: ``` ComfyUI/models/FloweR/FloweR_0.1.2.pth ComfyUI/models/RAFT/raft-things.pth ``` Or via CLI: ```bash huggingface-cli download pxlpshr/ComfyUI-SD-CN-Animation FloweR_0.1.2.pth --local-dir ComfyUI/models/FloweR/ huggingface-cli download pxlpshr/ComfyUI-SD-CN-Animation raft-things.pth --local-dir ComfyUI/models/RAFT/ ``` ## What This Enables ### Txt2Vid (Text to Video) Generate video sequences from a text prompt. FloweR drives the motion by predicting optical flow and the next frame, while Stable Diffusion adds detail and style through a two-pass sampling pipeline (inpaint + refine). Histogram matching prevents color drift. ### Vid2Vid (Video Stylization) Stylize existing videos with any SD checkpoint. RAFT computes real optical flow between input frames, warps the previous stylized frame forward, and uses occlusion detection to determine what needs regenerating. Maintains temporal coherence across the full sequence. Both modes support ControlNet and IPAdapter for additional control over style and structure. ## Model Origins - **FloweR** — Optical Flow Reconstruction model by [volotat](https://github.com/volotat), from the [SD-CN-Animation](https://github.com/volotat/SD-CN-Animation) project. - **RA...
Source context: 0 downloads · 1 likes · Pipeline image-to-video · Library comfyui · Repo pxlpshr/ComfyUI-SD-CN-Animation
