<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/43abe997-953d-4514-9787-c0100a523ddc/width=525/43abe997-953d-4514-9787-c0100a523ddc.jpeg" />Hi,this Style Lora inspirited by old "French Postcards" from 1900 till 1950. It can used for classic vintage erotic images or as only style Lora for all other your ideas..Text to Image works very well for me with the Euler beta with 9 - 11 steps.<ul><li>Prompt Tips: Start the Generation with "A vintage photo of.." or " A vintage postcard of...."</li><li>If you want black and white images avoid color description or use black and white for clothes, hair etc.</li><li>If you want monochrome style, use color description for the hair or sofa etc.</li><li>You can check my sample Images for prompting tips.</li></ul>It was trained with 150 classic Photos. in Version 2.0 i used the same training images but removed watermarks and text.

French Postcard Vintage Style

Hi, this is the version for the Moss Model. It based on the v2 z-Image Version.It works good with 8 steps (with the 8-steps-dmd2 lora by lodestones <a target="_blank" rel="ugc" href="https://huggingface.co/lodestones/Kromoss/tree/main">https://huggingface.co/lodestones/Kromoss/tree/main</a> ) ER_DE (simple) and Euler (simple).Lora strength is 1.0 for spicy Images, and 0.7 for not so spicy Images to use only the style.Is used the modified Basic moss text to Image comfyui workflow. The workflow are inside my sample Images.Some Tips from my testing: if you want black and white / monochrome images start the prompt with "a vintage black and white photo of..." and avoid color (like ginger hair etc) descriptions in the prompt. if you want a (re)colored style images, describe objects with colors in the prompt.

Files with SHA256: 4cdc4a88831b98851297a66face826beae80794b58c9012f226fc9fb91485e1d