---
title: Qwen-Image-Edit-Middle-Finger-LoRA
canonical_url: "https://www.modelscope.cn/models/drbaph/Qwen-Image-Edit-Middle-Finger-LoRA"
md_url: "https://www.modelscope.cn/models/drbaph/Qwen-Image-Edit-Middle-Finger-LoRA.md"
repository: drbaph/Qwen-Image-Edit-Middle-Finger-LoRA
last_updated: 2026-09-14
license: apache-2.0
pipeline_tag: image-to-image
tasks:
  - image-to-image
base_model:
  - Qwen/Qwen-Image-Edit
base_model_relation: adapter
parameters: 147.5M
tensor_type:
  - BF16
library_name:
  - lora
  - diffusers
  - safetensors
language:
  - en
supports_inference: img2img
downloads: 6
stars: 0
tags:
  - text-to-image
  - lora
  - diffusers
  - "template:diffusion-lora"
  - gesture
  - pose
---

# Qwen-Image-Edit-Middle-Finger-LoRA

> Qwen-Image-Edit-Middle-Finger-LoRA - drbaph 在 ModelScope 开源的模型。Qwen Image Edit Middle-Finger LoRA

drbaph/Qwen-Image-Edit-Middle-Finger-LoRA 是 ModelScope 魔搭社区上的 147.5M 参数image-to-image模型，采用 apache-2.0 许可，基于 Qwen/Qwen-Image-Edit 构建，并支持在线推理（img2img）。

- **Repository**: drbaph/Qwen-Image-Edit-Middle-Finger-LoRA
- **License**: apache-2.0
- **Tasks**: image-to-image
- **Parameters**: 147.5M
- **Base model**: Qwen/Qwen-Image-Edit
- **Online inference**: img2img
- **Tags**: text-to-image, lora, diffusers, template:diffusion-lora, gesture, pose
- **Downloads**: 6
- **Stars**: 0
- **Last updated**: 2026-09-14

Source: https://www.modelscope.cn/models/drbaph/Qwen-Image-Edit-Middle-Finger-LoRA

---

# Qwen Image Edit Middle-Finger LoRA

<Gallery />

## Model description 

# Qwen Image Edit Middle Finger LoRA

---

## 📌 Description
Its primary purpose is to **edit images** to **add, modify, or replace hand gestures**, specifically to make the subject **show a middle finger** when prompted.

---

## 📝 Usage
Use the following prompt:

    make this man putting middle finger up

For flexibility, you can replace **&quot;man&quot;** with:

    Man / Woman / Person / Character / Dog / Cat / Monkey / Chicken

---

## 🚀 Best Results
This model pairs well with the following **LoRAs** for faster and more refined results:

- **Qwen Image Lightning 4-step**
- **Qwen Image Lightning 8-step**

&gt; ⚡ These LoRAs are **optional** and only improve **speed**.  
&gt; The base model works fine without them.

---

## 🔄 Troubleshooting
- If the generated image **shows a pointing finger instead of a middle finger**, **change the seed** and try again.

---

## 🛠️ Training Information
- **Training Pairs:** 210  
- **Data Type:** AI-generated gesture-focused dataset  
- **Input:** Images with person &#x2F; character &#x2F; animal  
- **Output:** Edited images showing a **realistic middle finger gesture** based on learned patterns



## Download model


[Download](/drbaph/Qwen-Image-Edit-Middle-Finger-LoRA/tree/main) them in the Files & versions tab.