This is a mix based on <a rel="ugc" href="https://civarchive.com/models/257749/pony-diffusion-v6-xl">Pony Diffusion V6 XL. </a>During testing i got a more cinelike look in the images with a softer attenuated depth. Only a small fraction of the model is changed to not kill the Pony itself, but it seems, other LoRas beside the Pony ones work far better now. This model is a result from the try to make the Text Aid LoRa working for Pony. I tried mixes up to 50% where it started to work, but that fully killed the Pony already. Besides some minor other changes there is a fraction &lt;1% from <a rel="ugc" href="https://civarchive.com/models/133955/roxl">RoXL </a>included.

Files with SHA256: 14993ba89ba139683da309f5998093324b196e8e72f58dcfb7a81114e73f79f4