## 🧔 Middle-aged Truck Driver Bear Uncle —— Asian Middle-aged Exclusive中年司机大叔壮熊 —— 亚洲中年Design Concept: Focused on realistic Asian middle-aged truck driver bear daddies, emphasizing thick and heavy physique, life traces, and mature charm. This LoRA is specially fine-tuned to strengthen bear-style dad bod and authentic skin details, aiming to generate powerful yet warm and approachable real bear dad truck driver images, perfectly reproducing the thick, mature vibe from the reference photo.Core Features: Thick bear physique (thick chest, soft belly, broad shoulders, thick arms), Asian middle-aged facial features (realistic 45-60 years old, thick mustache, short black hair), steady truck driver aura with life details (skin texture, natural body hair, leather pants elements), mature and composed expressions.Main Advantages: Extremely realistic style, excellent thick body and soft belly rendering, stable hands and body proportions, rich natural expressions. Ideal for various daily/work scenes of Asian middle-aged bear uncles, covering 45-60 age range, highly compatible with Z-Image-turbo.### 🔑 Trigger Word *<code>hu2ne_nano</code>**- Compatible Base Model: Z-Image-turbo - Strengthens Asian middle-aged male features + bear physique + truck driver life feelRecommended Generation Settings: - Sampler: Euler - Steps:9 - CFG Scale: 1 - LoRA Weight: 0.8-1.0 (daily recommended 0.85, for stronger physique up to 1.0; for img2img or combining LoRAs, lower to 0.6-0.8) - Resolution: 1024x10244K Version Note: Some models support 4K dataset training, supports higher resolutions (recommended 1328x1328+), with better details and skin texture. Recommend using with Hires fix.### 💡 Practical Tips 1. Natural Language Prompt Recommendation: Best used with natural language supported models (like Z-Image-turbo). Use LLMs (ChatGPT, Claude, Gemini, Grok, Qwen) to polish prompts — simply describe the desired Asian middle-aged truck driver bear dad scene for more natural results. 2. Civitai Online Generation Note: Local generation recommended for best Z-Image-turbo performance. 3. Trigger Reminder: Use <code>hu2ne_nano</code> to activate core style. 4. Enhance Asian features → add <code>east asian features, black hair, monolids</code> at the end. 5. More bear/thick → add <code>prominent dad bod, thick waist, body hair, huge pecs, soft belly</code>. 6. Pose suggestions: <code>standing confidently, sitting in driver seat, relaxing on sofa</code>. 7. For better faces → add <code>highly detailed, 4K, ultra realistic, photorealistic, extremely detailed face</code> or lower LoRA weight.---Welcome to use and give feedback. If helpful, you can support me (BUZZ). Thanks!I organize LoRAs via Collection. For more Asian middle-aged bear / Bara Daddy content, check: 暂无

hu2ne_nano   Bear Uncle

zit_hu2ne_nano_v1_c1-st6000.safetensors

v1.0 (ZImageTurbo)


# 

## 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-Turbo](https://modelscope.ai/models/Tongyi-MAI/Z-Image-Turbo)
* Training Code: [DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
* Data amount：20
* Total training steps：6000
* 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-Turbo", origin_file_pattern="transformer/*.safetensors"),
        ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="text_encoder/*.safetensors"),
        ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
    ],
    tokenizer_config=ModelConfig(model_id="Tongyi-MAI/Z-Image-Turbo", origin_file_pattern="tokenizer/"),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="jlhe2016/zit_hu2ne_nano_v1", origin_file_pattern="zit_hu2ne_nano_v1_c1-st6000.safetensors"))
prompt = "a cat"
image = pipe(prompt=prompt, num_inference_steps=8, cfg_scale=1)
image.save("image.jpg")
```


Files with SHA256: 32fb45d11c732f95078c621d6ca1ddf99e2967910f3535a28e7cdaec27539c7f