❤️ Please Consider Supporting My Work
I love building open tools and LoRA models for everyone. If you find this useful, consider supporting me with a coffee. It helps me create even better and more models!
👉 https://ko-fi.com/vizsumit
Prompt: pencil sektch, maid, maid costume,
Checkpoint: JugglerNautXL
Description
FAQ
Comments (5)
I made a video sequence using this LORA and img2img, basically as a style transfer, using an image sequence generated from my webcam. It can be seen 33 seconds into this video (and runs for about 13 minutes -- it was an 8000 frame sequence and runs at 10fps). The prompt was "withLora(Pencil_Sketch_by_vizsumit,0.6) person sketch", and the image_strength value was 0.13, and the sampler was k_euler_a. https://youtu.be/sILcfGESI44?t=33
Hello, it seems like "pencil sketch" is not just a helpful prompt, but a required trigger word. I can't get any LoRA effect w/o it.
depends on size of your prompt and model you are using... give me more clues
@vizsumit revising my "any" to much, posting a picture.
Using img2img and when I use this, it makes my image uglier plus it doesn't even come out like a pencil sketch. It just regenerated the image.
Steps: 20
CFG: 7-9
Denoise: 0.3
Image size: W-341 x H-512
Prompt: a pencil sketch of a beautiful 18 year old woman in a pink bikini, <lora:Pencil_Sketch_by_vizsumit:1>
Any ideas?