Explore my workflow creation space! Feel free to click the link below and experience the future of AIGC generation directly.<a target="_blank" rel="ugc" href="https://www.runninghub.ai?inviteCode=rh-v1559">https://www.runninghub.ai?inviteCode=rh-v1559</a>Fan benefits,register to get 1000 points,daily login 100 points,play 4090!Experience the super power of 48G.## Model OverviewWelcome to Krea2_MegaCity_Greeble!A heavy-duty architecture and scenery LoRA tailored for Mega-Structure Cyberpunk Cityscapes, hyper-dense futuristic architecture, intricate greeble line art, and breathtaking multi-layered spatial perspectives.Inspired by classic sci-fi manga masters, this model excels at rendering vast dystopian cities overflowing with structural complexities. It effortlessly generates colossal skybridges, deeply layered skyscraper canyons, and crisp industrial line work with rich metallic shadows. An essential tool for sci-fi concept art, game worldbuilding, matte painting backgrounds, and high-tech wallpaper generation.## ⚙️ Recommended Parameters (Optimized for Krea2 Turbo)To capture the massive architectural scale and ultra-fine greeble textures, please apply the following configurations:* Base Architecture: Krea2* **Inference Pipeline:** Krea2 Turbo Version* Sampling Steps: 8 steps (Strictly optimized for Turbo speed)* CFG Scale: 1.0 (Crucial for Turbo models)* LoRA Weight: 0.8 - 1.0 (0.9 is highly recommended for optimal mechanical detail)* Core Trigger Tags: <code>megastructure style</code>, <code>cyberpunk cityscape</code>, <code>high density futuristic city</code>, <code>greeble details</code>, <code>intricate line art</code>, <code>multi-level skybridges</code>, <code>isometric view</code>## 🎨 Prompting Guide* To boost massive scale and panoramic verticality: <code>massive skybridges crossing between towers</code>, <code>extreme high angle view</code>, <code>endless vertical cityscape</code>, <code>deep abyss below buildings</code>* To enhance line art precision and industrial vibes: <code>dense greeble details</code>, <code>industrial mechanical textures</code>, <code>desaturated grey palette</code>, <code>detailed linework</code>

Krea2_CyberStructure_Isometric

Krea2 Turbo image.json

Krea2_CyberStructure_Isometric.safetensors

v1.0


# Krea2_科幻巨构繁筑器

## 模型介绍

本模型依托魔搭社区（ModelScope）AIGC专区[模型训练](https://modelscope.cn/aigc/modelTraining)环境与算力完成训练。

* 模型类型：LoRA
* 基础模型：[krea/Krea-2-Turbo](https://modelscope.cn/models/krea/Krea-2-Turbo)
* 训练代码：[DiffSynth-Studio](https://github.com/modelscope/DiffSynth-Studio)
* 训练数据量：100
* 总训练步数：10000
* 开源协议：Apache-2.0

## 推理代码

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

```bash
pip install diffsynth
```

开始推理：

```python
from diffsynth.pipelines.krea2 import Krea2Pipeline, ModelConfig
import torch

pipe = Krea2Pipeline.from_pretrained(
    torch_dtype=torch.bfloat16,
    device="cuda",
    model_configs=[
        ModelConfig(model_id="krea/Krea-2-Turbo", origin_file_pattern="turbo.safetensors"),
        ModelConfig(model_id="Qwen/Qwen3-VL-4B-Instruct", origin_file_pattern="*.safetensors"),
        ModelConfig(model_id="Qwen/Qwen-Image", origin_file_pattern="vae/diffusion_pytorch_model.safetensors"),
    ],
    tokenizer_config=ModelConfig(model_id="Qwen/Qwen3-VL-4B-Instruct", origin_file_pattern=""),
)
pipe.load_lora(pipe.dit, ModelConfig(model_id="zhouwenbin1994/Krea2_CyberStructure_Isometric", origin_file_pattern="Krea2_CyberStructure_Isometric_c1-st10000.safetensors"))
prompt = "a cat"
image = pipe(
    prompt, seed=0,
    num_inference_steps=8, cfg_scale=1, mu=1.15,
)
image.save("image.jpg")
```



Files with SHA256: 1d91adcab93fe69a7c5edc39606054863a0b3cb43cd27464d8b48038e8ee0f87