## 🧔 Middle-aged Muscular Bear Fitness Coach —— Asian Middle-aged 中年壮熊健身教练 —— 亚洲中年 触发词<code>lei nano</code> 核心：厚实肌肉熊体型、亚洲中年硬朗五官、健身教练气场 兼容：Flux Klein / Z-Image### Design Concept Focused on realistic muscle bear uncles, this LoRA specializes in depicting the heavy, powerful charm of Asian middle-aged men as fitness coaches. Through targeted fine-tuning, it strengthens the robust bear physique and lifelike details, aiming to generate authentic Asian bear daddy fitness coach images that combine explosive muscular strength with mature warmth.### Core Features Thick muscular bear physique (massive pecs, thick arms, powerful thighs, visible abs), Asian middle-aged hard facial features (short black hair, stubble, steady and serious expression), professional fitness coach aura, visible body hair, realistic skin texture with sweat &amp; vein details.### Main Strengths Extreme realistic muscle details, excellent thick bear body performance, stable hand &amp; body proportions, natural fitness scenes, authentic age feel (45-55 mature male). Ideal for gym/training scenes and daily bear daddy lifestyle shots with strong muscular presence and accurate Asian features.### 🔑 Trigger Word *<code>lei nano</code>**- Compatible Base Models: Flux Klein / Z-Image - Strengthens Asian middle-aged male features + Bear muscular physique### 📋 Recommended Prompt Examples1. Fitness Coach Daddy a powerful mature Asian bear daddy fitness coach in his late 40s-50s, massive thick chest and broad shoulders, huge muscular arms and thick thighs, prominent abs and body hair, short black hair with stubble, confident serious expression looking at viewer, standing in gym with hands on hips, detailed realistic skin texture with sweat and veins, masculine and dominant, soft dramatic lighting, highly detailed, photorealistic2. Sweaty Training Bear Dad a sweaty muscular Asian bear daddy fitness coach in his early 50s, wearing tight black gym shorts, enormous pecs and thick waist, veiny forearms and huge legs, body hair visible on chest and abs, intense focused expression, mid-workout pose, gym background with equipment, cinematic warm lighting, ultra realistic skin and muscle details, highly detailed3. Relaxed Muscle Bear Coach (Intimate Feel) a strong yet approachable Asian bear daddy in his mid 50s after workout, shirtless with only black shorts, massive dad bod muscles with soft belly layer over hard abs, broad shoulders and thick neck, gentle smile with sweat glistening on skin, relaxed powerful standing pose, soft warm gym lighting, extremely detailed realistic face and body texture, very masculine and attractive### Recommended Generation Settings - Sampler: Euler - Steps: 20-30 - CFG Scale: 4-5 - LoRA Strength: 0.8-1.0 (daily recommended 0.85; stronger muscles → 1.0; editing/combining with other LoRAs → 0.6-0.8) - Resolution: 1024x1024 (supports up to 1328x1328+ for 4K dataset version)4K Version Note: This model supports 4K training data — higher resolutions will show superior muscle texture and skin details. Recommend using with Hires fix.### 💡 Practical Tips 1. Best used with natural language prompts on Flux Klein / Z-Image. Let ChatGPT / Claude / Grok expand your scene description. 2. Local generation strongly recommended (Civitai online may not trigger Z-Image LoRA well). 3. Enhance Asian look: add <code>east asian features, black hair, monolids</code> at the end. 4. More bear/thick: add <code>prominent dad bod, thick waist, body hair, huge pecs, massive thighs</code>. 5. Pose ideas: <code>hands on hips, flexing biceps, mid-workout, standing confidently</code>.---Collection：暂无

Middle-aged Muscular Bear Fitness Coach    中年壮熊健身教练

zimageturbo-uiu23_c1-st2000.safetensors

v2.0(Zimage turbo ZIT)

v1.0(flux.2 klein 9B)

v1.0（Z image Base ZIB）


# 

## 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/zimageturbo-uiu23", origin_file_pattern="zimageturbo-uiu23_c1-st6000.safetensors"))
prompt = "a cat"
image = pipe(prompt=prompt, num_inference_steps=8, cfg_scale=1)
image.save("image.jpg")
```


Files with SHA256: b9a707b8e1da8605a0148433a196263cfce76a3c1d34d8daeca80a73280f5e93