Mila — Multi-Model Character LoRAs

Mila is a fictional AI character developed for consistent identity generation across multiple image and video model architectures.

This repository contains dedicated Mila LoRA weights for:

MiniMax H3 · LTX-2.3 · Z-Image Turbo · Krea 2 Turbo

Each LoRA is trained specifically for its corresponding base model and should only be used with that architecture.

Prompt: Mila — realistic cafe portrait

Prompt: Mila — Commercial video

Prompt: Mila — Skincare Commercial

Prompt: Mila — Santa Monica Walk

Prompt: Mila — Grocery shopping video

Prompt: Mila — luxury car video

Example Prompt

mila_v1, a blonde adult woman with softly defined facial features, natural skin texture and loose strands framing her face.
Seated at an outdoor cafe table, wearing a fitted black scoop-neck top, a small pendant and a wristwatch, with her blonde hair tied back.
She glances downward, lifts her face, smiles broadly and raises a hand beside her hair.
A white cup rests near the table. Plants and warm restaurant lights form a softly blurred background.
sound: silent video; no audible speech, music or ambient sound.

About Mila

Mila is a fictional AI-generated character.

The project explores consistent character identity across independent generative model architectures while preserving flexibility in:

Clothing
Hairstyle
Pose
Expression
Environment
Lighting
Camera framing
T2V and I2V generation

Rather than locking the character to a single scene or aesthetic, the LoRAs are intended to preserve Mila's recognizable identity across a wide range of realistic scenarios.