Please note - this model already includes CLIP T5 and L, so you only need to load it directly from the checkpoint if using ComfyUI
This model is still in the works and going through some updates and testing. Sometimes the results might not be spot-on because the dataset is still pretty limited
Recommend resolution : 768 x 1024
- No trigger words are needed
- Note: I haven't experimented with other LoRAs yet, so the weights may require further adjustment.
Try online https://www.shakker.ai/userpage/179558d95f804022b326701c32618dd4
Description
FAQ
Comments (1)
Looks neat, mate!
But can you please separate uNet with other stuff?
Also, I think that Q6_K_M quant will be a good addition. Usually, it is pretty similar to Q8.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.