However, that's another story altogether! Also check the model with fewer steps <a target="_blank" rel="ugc" href="https://civarchive.com/models/43735">https://civarchive.com/models/43735</a>, in some cases it may be more flexible. As these both LoRAs were trained in kinda same way - primary model was deliberate - don't forget about it's weight - as was mentioned in that lora, it works well when the value is around 0.6 - 0.8. At the end, if you are also using deliberate_v2, I recommend reducing the CFG value to 3.5 - 5.

Files with SHA256: a7183199592cf79b0741944a6c0ff925e008c30392a32214155d8f108541b62a