It's a LoRA trained on Kiara and Kovu from Disney The Lion King 2. I recommend use the LoRA weight between 0.7-1. Better works on animal and furry models.
To get the result you need use lora and write Kiaraandkovu, Kiara_tlk, Kovu_tlk.
The model works good with both Kiara and Kovu (use something like: 1 lion 1 lioness or lion and lioness, together) and with Kovu himself. It doesn't generates images with Kiara without Kovu at all, I don't know why.
Also you can use Kiara_tlk Kovu_tlk cuddling or Kiara_tlk Kovu_tlk in Upendi Heart.
If you wanna get images when Kovu with scar you can write (Kovu_tlk with scar) I don't know how correctly it will work because I don't like him with scar so I don't check it much, but the model was trained a little for that too. However I don't guarantee that it will work.
The LoRA trained on 768x512 images.
Description
Trained on 768x512 images
FAQ
Comments (3)
What model diffusion did you use?
What do you mean? All data is available in posts. If you ask about training checkpoint, it was trained on basic 1.5 version.
This just doesn't work. I get cheetahs, cougars, huskies, bears... Tried across a dozen different models, various weightings, aggressive forced tagging, and still not a single Kovu and Kiara. :(
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.