This model combines all of my other models based on ground-up retraining from the source material and additional mergers with the other models to get an even and balanced result that allows a combination of all of them. There is a LoRA version. See Link |