Separate LLM that rasterizes latent image. Although given that you're using service instead of proper UI, I'd assume you have no control over such parameters and instead have very high CFG scale, poor weight scaling, or bad LoRA or checkpoint outright. In other words - skill issue.
Separate LLM that rasterizes latent image. Although given that you're using service instead of proper UI, I'd assume you have no control over such parameters and instead have very high CFG scale, poor weight scaling, or bad LoRA or checkpoint outright. In other words - skill issue.
A VAE isn't a LLM, it's an encoder. It's in the name.
hello! and how to add a more realistic nose like in the previous model?
@UmiZhigalo 1. Add the "realistic" tag, although this also slightly changes the style in general. 2. Use the previous model. Completely or possibly use it as an inpaint on the face. 3. Wait. Perhaps in the next versions I will be able to bring the style closer to what it was before. I myself like that style more, but the new model is more obedient and less prone to anatomy defects.
Good gods, that's a dangerous looking weapon. (and there's something wrong around the right pov hand, but I'm not going to flag it because of the sheer magnificence of the image)
