Detail enhancer for hair. Adds strand separation, flyaway hairs and braid structure across all hair types.Trigger word: sdrhair01WHAT IT DOESBase models tend to render hair as a smooth mass, especially in backlight, wet hair and tight braid patterns. This LoRA pushes individual strands to separate and holds structure in weaves and partings.Trained on 70 photographs covering box braids, knotless braids, cornrows, locs, straight, wavy, curly, ponytails and wet hair.HOW TO USEPut sdrhair01 at the start of your prompt.Recommended strength: 0.8 to 1.0.Describe hair colour explicitly, otherwise the model may pull towards darker tones.WHAT IT IS NOT GOOD ATTrained on female subjects only. Results on male hair are unpredictable.Short cuts are underrepresented in the dataset.Base model: Z ImageFree to use. Feedback and sample images are welcome.Made by SEEDRA<a target="_blank" rel="ugc" href="http://seedraai.com">seedraai.com</a> Our Discord Server: <a target="_blank" rel="ugc" href="https://discord.gg/seedraai">https://discord.gg/seedraai</a>

SEEDRA Hair Detail — sdrhair01

SeedraAI_Hair_c1-st2000.safetensors

v1.0


# 

## Model Introduction

This model is trained using the [Civision/model training](https://modelscope.ai/civision/modelTraining) environment and computing resources provided by ModelScope.

* Model Type: LoRA
* Base Model: [Tongyi-MAI/Z-Image](https://modelscope.ai/models/Tongyi-MAI/Z-Image)
* Training Code: [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
* Data amount：57
* Total training steps：2000
* License: Apache-2.0

## Inference Code

Install [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio):

```bash
pip install diffsynth
```

Inference:

```python
from diffsynth.pipelines.z_image import ZImagePipeline, ModelConfig
import torch

pipe = ZImagePipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="Tongyi-MAI/Z-Image", origin_file_pattern="transformer/*.safetensors"),
        ModelConfig(model_id="Tongyi-MAI/Z-Image", origin_file_pattern="text_encoder/*.safetensors"),
        ModelConfig(model_id="Tongyi-MAI/Z-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
    ],
    tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image", origin_file_pattern="tokenizer/"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="cfgscale/SeedraAI_Hair", origin_file_pattern="SeedraAI_Hair_c1-st2000.safetensors"))
prompt = "a cat"
image = pipe(prompt=prompt, num_inference_steps=50, cfg_scale=4)
image.save("image.jpg")
```


Files with SHA256: 685b88918cddb9466a8e1af9c8129143ed5f0d0b9c39ce454bf8d7f693a85258