MJ7 VERSION Update: The dataset was rebuilt entirely from scratch using MidJourney V7.The main difference from v1 and v2 is a shift towards more natural colors and textures. The Lora is trained on MidJourney Images that use my "Personalised Style" + Watercolor What is new in v2:<ul><li>Significantly increased the training dataset</li><li>Added support for a wider range of concepts</li><li>Removed most of the biases that were present in v1 (closed eyes everywhere for example)</li><li>Added 3 different intensity of the colors ( intense , deep, soft colors)</li><li>My personal style is changed a little bit , so the images are more "crisp and clear"</li></ul>Works without the trigger words ArsMJStyle, WatercolorWorks great in the 0.2 - 1.5 range.Start ~1.As Watercolor is strong existing style and looks really nice as an abstract painting, you can go pretty high with the strength and still get a nice result.I hope you will enjoy this LoRA as much as I did creating it!

Ars MidJourney Watercolor

The training images for this version were the same as for V2 of the Pony/SDXL , so everything from model description apply here as well. If you use quantized version use slightly higher strength values. The color keywords (intense , deep, soft colors) work much better with Flux , which was a pleasant surprise.For recommended usage on Quantized version check this really simple ComfyUI Workflow:<a target="_blank" rel="ugc" href="https://civitai.com/models/658639/super-simple-gguf-quantized-flux-lora-workflow">https://civitai.com/models/658639/super-simple-gguf-quantized-flux-lora-workflow</a>All the images with missing Metadata from the Showcase Gallery were created with it and Flux Q4 Dev model.I really enjoyed creating and testing this, so I hope you will enjoy using it as well 😁

Files with SHA256: 71e70e43eea7d50506fe47970999605fe606642ace27360e266f53dc2c568654