If you'd like to support me then why not check out my DeviantArt here you can support me by buying a premium download or buying any of my premium galleries. I updated twice a week and use a variety of different styles. So if you like what you see then head on over. <a target="_blank" rel="ugc" href="https://www.deviantart.com/consistentfactorcf">https://www.deviantart.com/consistentfactorcf</a>OrchidXL is another model that is part trained, this one is made to have a slight watercolour style to it, it was very difficult to get this to stick so sadly this version requires a trigger word which is watercolor. I first trained 100 images to make a lora and merged it into the model, maybe next time I'll use more images to try and not have to use a trigger word.All my SDXL models come with this VAE baked in <a target="_blank" rel="ugc" href="https://civarchive.com/models/140686/fix-fp16-errors-sdxl-lower-memory-use-sdxl-vae-fp16-fix-by-madebyollin?modelVersionId=155933">https://civarchive.com/models/140686/fix-fp16-errors-sdxl-lower-memory-use-sdxl-vae-fp16-fix-by-madebyollin?modelVersionId=155933</a>

OrchidXL

Initial version part trained on a 100 image lora then merged.

Files with SHA256: 33d3f95024b1b4850cd1f55b2143c10c18fae090a7c2027db11d339c87f8d9f4