<h2 id="metrorose-realism-female-lora">🌹 MetroRose — Realism Female LoRA</h2><p><strong>MetroRose</strong> is a realism-focused female LoRA designed to create natural-looking, high-quality AI-generated women with detailed facial features, realistic skin, and photographic aesthetics.</p><p>Great for portraits, fashion, lifestyle, street photography, and cinematic realism.</p><h3 id="settings:-flux.2">✨ Settings: Flux.2</h3><ul><li><p>Strength: 0.4-0.6</p></li><li><p>Clip Skip: 2</p></li><li><p>CFG: 4-6</p></li><li><p>Upscale: UltraSharp 4x</p></li><li><p>Sampler: DPM 2</p></li></ul><p></p><h3 id="settings:-z-image-base-and-turbo">✨ Settings: Z-Image Base &amp; Turbo</h3><ul><li><p>Strength: 0.7-0.9</p></li><li><p>Steps: 12 ( Turbo) - 24 (Base)</p></li><li><p>Clip Skip: 2</p></li><li><p>CFG: 4-6</p></li><li><p>Upscale: UltraSharp 4x</p></li><li><p>Sampler: Euler</p></li></ul><h2 id="wan-video-2.1-14b-version:-https:www.patreon.comtensorvizionpostsmetrorose-video-166713562">Wan Video 2.1 14B Version: <a target="_blank" rel="ugc" href="https://www.patreon.com/TensorVizion/posts/metrorose-video-166713562">https://www.patreon.com/TensorVizion/posts/metrorose-video-166713562</a></h2><h2 id="krea-2-version:-https:www.patreon.comtensorvizionpostsmetrorose-krea-2-166552675">Krea 2 Version: <a target="_blank" rel="ugc" href="https://www.patreon.com/TensorVizion/posts/metrorose-krea-2-166552675">https://www.patreon.com/TensorVizion/posts/metrorose-krea-2-166552675</a></h2><h3 id="trigger">🏷️ Trigger</h3><p><code>Cherry</code></p><h3 id="ai-disclosure">📸 AI Disclosure</h3><p><strong>All images shown on this model page are fully AI-generated.</strong> They are not photographs of real people and are intended solely as demonstrations of the LoRA's capabilities.</p><p><strong>Have fun creating! 🌹</strong></p><p><strong>Made using AIGC Studio Civision</strong></p>

MetroRose

<p><strong><u>Trained on HiDream</u></strong></p><p>Shout Out to DiffSynth Studio for the Training Script</p>

MetroRose-HiDream_c1-st5000.safetensors

HiDream

Event Horizon XL

Illustrious

Flux.2 Klein 4B

Flux.2


# 

## 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: [HiDream-ai/HiDream-O1-Image](https://modelscope.ai/models/HiDream-ai/HiDream-O1-Image)
* Training Code: [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
* Data amount：52
* Total training steps：5000
* License: Apache-2.0

## Inference Code

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

```bash
pip install diffsynth
```

Inference:

```python
from diffsynth.pipelines.hidream_o1_image import HiDreamO1ImagePipeline, ModelConfig
import torch

pipe = HiDreamO1ImagePipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="HiDream-ai/HiDream-O1-Image", origin_file_pattern="*.safetensors"),
    ],
    processor_config=ModelConfig(model_id="HiDream-ai/HiDream-O1-Image"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="tensorvizion/MetroRose-HiDream", origin_file_pattern="MetroRose-HiDream_c1-st5000.safetensors"))
prompt = "a cat"
image = pipe(prompt=prompt, num_inference_steps=30, cfg_scale=4)
image.save("image.jpg")
```


Files with SHA256: 26e5bb59d6a56405457166a91d3d7c8967e11ee10d1c6f17141bdff18581d0f7