nodes/city96-ComfyUI_ExtraModels
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Update README.md

Browse Source 
#84
This commit is contained in:
City
2024-12-05 19:19:37 +01:00
committed by GitHub
parent e9cc31c9cb
commit 78e10dab45
1 changed files with 24 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+24 -3
View File
@@ -36,6 +36,27 @@ git pull
Alternatively, use the manager, assuming it has an update function.
## Sana
[Original Repo](https://github.com/NVlabs/Sana)
### Model info / implementation
- Uses Gemma2 2B as the text encoder
- Multiple resolutions and models available
- Compressed latent space (32 channels, /32 compression) - needs custom VAE
### Usage
1. Download the model weights from the [Sana HF repo](https://huggingface.co/Efficient-Large-Model/Sana_1600M_1024px/tree/main/checkpoints) - the HF account has alternative models available too.
2. Place them in your checkpoints folder
3. Load them with the correct PixArt checkpoint loader
4. Use the "Gemma Loader" node - it should automatically download the requested model from Huggingface - Recommended to use the 4bit quantized model on CPU when low on memory.
5. Download the VAE from [here](https://huggingface.co/Efficient-Large-Model/dc_ae_f32c32_sana_1.0_diffusers/tree/main) and place it in your VAE folder after renaming it.
6. Use either the "Empty Sana Latent Image" or "Empty DCAE Latent Image" node for the latent input when doing txt2img.
[Sample workflow](https://github.com/user-attachments/files/18027854/SanaV1.json)
![Sana](https://github.com/user-attachments/assets/e4334352-e894-416a-b200-dff096027481)
## PixArt
@@ -50,9 +71,9 @@ Alternatively, use the manager, assuming it has an update function.
### Usage
1. Download the model weights from the [PixArt alpha repo](https://huggingface.co/PixArt-alpha/PixArt-alpha/tree/main) - you most likely want the 1024px one - `PixArt-XL-2-1024-MS.pth`
2. Place them in your checkpoints folder
3. Load them with the correct PixArt checkpoint loader
4. **Follow the T5v11 section of this readme** to set up the T5 text encoder
3. Place them in your checkpoints folder
4. Load them with the correct PixArt checkpoint loader
5. **Follow the T5v11 section of this readme** to set up the T5 text encoder
> [!TIP]
> You should be able to use the model with the default KSampler if you're on the latest version of the node.