Pooja is an original, entirely AI-generated adult Indian kemonomimi character. This Illustrious-trained character LoRA focuses primarily on her facial appearance, with long dark brown hair, a center part, striking green eyes, warm tan skin, and rounded black leopard ears defining her recognizable look.
Use it to bring Pooja into portraits, character illustrations, and wider scenes while allowing your checkpoint and prompt to shape the clothing, setting, and overall presentation.
Quick start
Trigger word:
PoojaFaceSuggested starting strength: 0.5
Suggested range to explore: 0.5–0.75
Training base: Illustrious
Primary focus: Facial identity and recognizable character features
Start at 0.5 and adjust gradually. These are suggested starting settings rather than a verified optimal range for Pooja. The best strength will depend on your checkpoint, prompt, and other resources. Higher strength is not necessarily better for likeness or image quality.
Checkpoint compatibility
Use an Illustrious-compatible checkpoint suited to the rendering style you want. Semi-realistic, stylized 3D/CG, and anime interpretations are useful styles to explore.
Pooja’s training imagery combines photographic settings and textures with stylized facial features. Some of that rendering may carry into your results. The balance between character likeness and your checkpoint’s style will vary with LoRA strength.
What it is designed for
Pooja is primarily a facial-identity LoRA, with a dataset covering front views, three-quarter angles, profiles, elevated and low viewpoints, and varied expressions.
The dataset also includes standing, seated, walking, bending, reclining, and hands-and-knees poses. Clothing and backgrounds vary, so the character is not tied to one specific outfit or setting.
Close-up and upper-body portraits are good starting points for evaluating likeness. Wider compositions are represented in the training imagery, but the LoRA is not intended to precisely reproduce a particular body, outfit, or background.
Hair and leopard features
Pooja’s long, center-parted dark brown hair and black leopard ears are recurring features throughout the dataset. These may appear readily even when they are not explicitly requested.
Her design includes human ears as well as furred animal ears. The intended leopard ears have a rounded shape, though their silhouette varies somewhat across the training images.
A long, thick black leopard tail with subtle rosette markings appears in many wider compositions. To reinforce the intended species, describe rounded black leopard ears and a black leopard tail with subtle rosettes, rather than relying only on a general “catgirl” prompt.
Hairstyle changes and the removal of animal features may require experimentation. Treat this as a Pooja character LoRA rather than a face-only tool that leaves every other head feature untouched.
Things to watch for
Distant faces: Small faces provide less room for eye alignment and fine facial details to resolve.
Expressive mouths: Smiles, laughter, surprise, and speaking expressions are represented in the dataset, but teeth, mouth shape, and likeness still warrant inspection.
Hair and ears: These features may remain associated with the character when you try to change them. Check that the ears retain the rounded leopard appearance you want.
Tails: Check tail count, attachment, markings, and interaction with clothing or furniture in wider compositions.
Strength and style: Stronger settings may carry more of the training imagery’s appearance into the result.
These are areas to evaluate in your workflow rather than fixed failures at particular settings. For wider compositions, a restrained face-detailing or inpainting pass may help.
Prompting
Include PoojaFace, then describe the expression, framing, clothing, lighting, and setting you want. Begin with a straightforward prompt and the LoRA at 0.5 before adding other resources or more complex instructions.
For example:
PoojaFace, 1girl, solo, adult Indian woman, portrait, looking at viewer, closed-mouth smile, tan skin, striking green eyes, long dark brown hair, center part, rounded black leopard ears, cream embroidered blouse, soft lighting, simple background
For wider compositions, add:
black leopard tail, subtle rosettes on tail
Adjust the style and quality terms to suit your chosen checkpoint. For expressions, describe the visible result—such as “closed-mouth smile,” “raised eyebrows,” or “slightly parted lips”—rather than relying only on a broad mood label.
Character and source
Pooja is a fictional adult Indian character created using AI-generated imagery. She is not intended to represent a real person.
Feedback
If you share results or report an issue, please include your checkpoint and version, LoRA strength, prompt, and any face-detailing or upscaling steps. That makes it easier to compare results and understand what works well.



