---
title: Qwen-Edit-3DChibi-LoRA
canonical_url: "https://www.modelscope.cn/models/AI-ModelScope/Qwen-Edit-3DChibi-LoRA"
md_url: "https://www.modelscope.cn/models/AI-ModelScope/Qwen-Edit-3DChibi-LoRA.md"
repository: AI-ModelScope/Qwen-Edit-3DChibi-LoRA
last_updated: 2026-06-17
license: apache-2.0
pipeline_tag: image-to-image
tasks:
  - image-to-image
base_model:
  - Qwen/Qwen-Image-Edit-2509
base_model_relation: adapter
parameters: 294.9M
tensor_type:
  - BF16
library_name:
  - pytorch
  - lora
  - diffusers
  - safetensors
frameworks:
  - pytorch
supports_inference: img2img
downloads: 98
stars: 4
tags:
  - lora
  - diffusers
  - "template:diffusion-lora"
  - Image-to-Image
---

# Qwen-Edit-3DChibi-LoRA

> Qwen-Edit-3DChibi-LoRA - AI-ModelScope 在 ModelScope 开源的模型。Qwen-Edit-3DChibi-LoRA

AI-ModelScope/Qwen-Edit-3DChibi-LoRA 是 ModelScope 魔搭社区上的 294.9M 参数image-to-image模型，采用 apache-2.0 许可，基于 Qwen/Qwen-Image-Edit-2509 构建，并支持在线推理（img2img）。

- **Repository**: AI-ModelScope/Qwen-Edit-3DChibi-LoRA
- **License**: apache-2.0
- **Tasks**: image-to-image
- **Parameters**: 294.9M
- **Base model**: Qwen/Qwen-Image-Edit-2509
- **Online inference**: img2img
- **Tags**: lora, diffusers, template:diffusion-lora, Image-to-Image
- **Downloads**: 98
- **Stars**: 4
- **Last updated**: 2026-06-17

Source: https://www.modelscope.cn/models/AI-ModelScope/Qwen-Edit-3DChibi-LoRA

---

# Qwen-Edit-3DChibi-LoRA

<Gallery />

## Model description 
This is a Low-Rank Adaptation (LoRA) checkpoint trained on the powerful Qwen-Image-Edit-2509 base model. It specializes in generating high-quality, highly detailed 3D Chibi Style images.


## Inference Example

```python
import torch
from diffusers import QwenImageEditPlusPipeline
from diffusers.utils import load_image

pipeline = QwenImageEditPlusPipeline.from_pretrained(
    "Qwen/Qwen-Image-Edit-2509",
    torch_dtype=torch.bfloat16
)
pipeline.load_lora_weights(
    "rsshekhawat/Qwen-Edit-3DChibi-LoRA", 
    weight_name="qwen_3d_chibi_lora_v1_000000820.safetensors"
)
pipeline.enable_sequential_cpu_offload()
image = load_image("Image.png").convert("RGB")

generator = torch.Generator(device="cuda").manual_seed(42)

prompt = "Convert this image into 3D Chibi Style"
image = pipeline(
    image=image, 
    prompt=prompt, 
    negative_prompt="",
    height=1024, 
    width=1024, 
    num_inference_steps=25,
    true_cfg_scale=5.0, 
    generator=generator
).images[0]
image.save("Edited_Image.png")
```

## Trigger words

You should use `3D Chibi Style` to trigger the image generation.


## Download model


[Download](/rsshekhawat/Qwen-Edit-3DChibi-LoRA/tree/main) them in the Files & versions tab.