import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Tongyi-MAI/Z-Image-Turbo", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("hodgy/luna_zit")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Luna — Z-Image Turbo LoRA
An identity/character LoRA for Tongyi-MAI/Z-Image-Turbo, trained with Ostris AI Toolkit.
Suggested trigger word
Start prompts with luna. This word is inferred from the repository and filename. The embedded internal output name is my_first_lora_v1, which is not necessarily the caption token, and the original captions are not included.
Example prompt:
Photorealistic portrait of luna, direct gaze, natural window light, detailed face, realistic skin texture, 85mm photography.
Training metadata
| Parameter | Value |
|---|---|
| Base family | Z-Image |
| Training steps | 1000 |
| Recorded epoch | 18 |
| Trainer | ostris/ai-toolkit 0.11.2 |
| Tensor count | 480 |
Recommended settings
- Base model:
Tongyi-MAI/Z-Image-Turbo - Start with LoRA strength
0.8–1.0. - Use 8 inference steps.
- State pose, framing, wardrobe, lighting, and setting explicitly.
fal.ai example
{
"prompt": "Photorealistic portrait of luna, natural light, realistic skin texture, 85mm photography",
"image_size": "portrait_4_3",
"num_inference_steps": 8,
"loras": [{
"path": "https://huggingface.co/hodgy/luna_zit/resolve/main/luna_z-image-turbo_lora_v1.safetensors",
"scale": 0.9
}]
}
Limitations and responsible use
The dataset, learning rate, and exact caption token are not included. Results vary with pose, framing, occlusion, and style. Use only with appropriate rights and consent; do not use for impersonation, deception, harassment, exploitation, or privacy violations. Disclose AI-generated imagery where appropriate.
- Downloads last month
- 23
Model tree for hodgy/luna_zit
Base model
Tongyi-MAI/Z-Image-Turbo