LORA trained on <a rel="ugc" href="https://mgq.miraheze.org/wiki/Morrigan">Morrigan</a> from MGQ.Training based on ingame CG and fanart. Trained on <a target="_blank" rel="ugc" href="https://civarchive.com/models/10028/neverending-dream-ned">NED</a> and tested also with AOMv2 and <a target="_blank" rel="ugc" href="https://huggingface.co/andite/anything-v4.0/tree/main">anything-v4.5</a>.Would suggest lower weights around ~0.65.With weights above 0.75 it becomes difficult to obtain poses different from the in game CG (for future version gonna try add more variety by including good AI generated images from previous version).For better and more flexible results i would suggest using ~0.5 of this LORA in combination with ~0.3 of my <a rel="ugc" href="https://civarchive.com/models/23494/shiki-style">Shiki style LORA</a> (the artist that draws this character)

Files with SHA256: add9c6cd96ce43a70be38226174a2c3c8be07f35305e0dd7b8f778372df574ef