Sign In

Furry and Friends - Anthropomorphic creatures

58
685
861
33
Verified:
SafeTensor
Type
LoRA
Stats
492
596
2.4k
Reviews
Published
Sep 24, 2024
Base Model
Flux.1 D
Usage Tips
Strength: 0.9
Trigger Words
anthropomorphic female
anthropomorphic male
furry
anthro
Hash
AutoV2
F67EC07114
The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.
IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

Furry and Friends - Anthropomorphic Creatures


V2 out with improved fur and a wider range of styles.
- Improved anatomy (better digitigrade legs) and nudity

This LoRA is designed to create realistic anthropomorphic animal hybrids, blending human and animal features in a semirealistic digital art style. While the primary focus is on detailed fur and realistic facial features, it can also achieve a photorealistic look with the right prompts. The comic style with outlines wasn't heavily trained, as my goal was to emphasize the texture and appearance of animal fur.

How to Use:

  • To trigger the LoRA effectively, include the animal's name after the person or character in your prompt. Adding terms like "furry" or "anthro" can help enhance the desired effect.

  • Describing the fur, fur patterns, feathers, or scales in your prompts can significantly enhance the realism and detail of the generated images.

  • Keep in mind that not all animal species are fully developed yet, as they require more training data. The faces, in particular, aim for a strong blend of human and animal characteristics.

This LoRA was trained to the limits of my 24GB GPU using 1200 carefully selected images (many from my Realistic Feline Hybrid LoRA for SDXL). I’m continuing to work on improving this LoRA and have several ideas for enhancements.

I welcome feedback on which animal species you'd like to see better represented, so I can focus on those in future updates.