modelscope-zimage-generator

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

ModelScope Z-Image Generator

ModelScope Z-Image 生成器

Generate images using ModelScope's Z-Image series models.

使用ModelScope的Z-Image系列模型生成图片。

IMPORTANT: Response Format

重要提示：响应格式

After generating an image, you MUST:

Use the Python script to generate the image
Respond with ONLY a simple text message like: "Image saved to: /path/to/output.jpg"
DO NOT include the image URL in your response
DO NOT try to display or reference the image URL - this will cause errors with non-multimodal models

The user can view the generated image at the file path you provide.

生成图片后，你必须：

使用Python脚本生成图片
仅回复简单的文本消息，例如："图片已保存至：/path/to/output.jpg"
请勿在回复中包含图片URL
请勿尝试展示或引用图片URL - 这会导致非多模态模型出错

用户可以通过你提供的文件路径查看生成的图片。

Prerequisites

前置条件

Get your ModelScope API key from: https://modelscope.cn/my/myaccesstoken

The script will automatically prompt for your API key the first time you use it and offer to save it for future use.

Optional: Manual Configuration

You can also manually configure the API key:

Config file:

bash

mkdir -p ~/.config/modelscope
cat > ~/.config/modelscope/config.json << EOF
{
  "api_key": "ms-your-api-key-here"
}
EOF

Environment variable:

bash

export MODELSCOPE_API_KEY="ms-your-api-key-here"

从以下地址获取你的ModelScope API密钥：https://modelscope.cn/my/myaccesstoken

首次使用脚本时，它会自动提示你输入API密钥，并提供保存密钥以便后续使用的选项。

可选：手动配置

你也可以手动配置API密钥：

配置文件：

bash

mkdir -p ~/.config/modelscope
cat > ~/.config/modelscope/config.json << EOF
{
  "api_key": "ms-your-api-key-here"
}
EOF

环境变量：

bash

export MODELSCOPE_API_KEY="ms-your-api-key-here"

Quick Start

快速开始

Use the Python script to generate images:

bash

cd /Users/ningoo/.claude/skills/modelscope-image-generator/scripts
python generate_image.py "A golden cat" output.jpg

使用Python脚本生成图片：

bash

cd /Users/ningoo/.claude/skills/modelscope-image-generator/scripts
python generate_image.py "一只金色的猫" output.jpg

Models

模型

Available Z-Image models:

```
Tongyi-MAI/Z-Image-Turbo
```
(default, fast generation)
```
Tongyi-MAI/Z-Image
```
(balanced quality and speed)
```
Tongyi-MAI/Z-Image-Edit
```
(image editing)

可用的Z-Image模型：

```
Tongyi-MAI/Z-Image-Turbo
```
（默认，生成速度快）
```
Tongyi-MAI/Z-Image
```
（画质与速度平衡）
```
Tongyi-MAI/Z-Image-Edit
```
（图片编辑）

Model Selection Rule

模型选择规则

If the user explicitly mentions "Z-Image-Turbo", use
```
Tongyi-MAI/Z-Image-Turbo
```
If the user explicitly mentions "Z-Image" (without Turbo), use
```
Tongyi-MAI/Z-Image
```
Otherwise, use the default
```
Tongyi-MAI/Z-Image-Turbo
```

Specify a different model using

--model

bash

python generate_image.py "A golden cat" output.jpg --model "Tongyi-MAI/Z-Image-Turbo"

如果用户明确提到"Z-Image-Turbo"，使用
```
Tongyi-MAI/Z-Image-Turbo
```
如果用户明确提到"Z-Image"（不带Turbo），使用
```
Tongyi-MAI/Z-Image
```
否则，使用默认的
```
Tongyi-MAI/Z-Image-Turbo
```

使用

--model

参数指定其他模型：

bash

python generate_image.py "一只金色的猫" output.jpg --model "Tongyi-MAI/Z-Image-Turbo"

LoRA Support

LoRA支持

Single LoRA:

python

"loras": "<lora-repo-id>"

Multiple LoRAs (weights must sum to 1.0):

python

"loras": {"<lora-id1>": 0.6, "<lora-id2>": 0.4}

单个LoRA：

python

"loras": "<lora-repo-id>"

多个LoRA（权重总和必须为1.0）：

python

"loras": {"<lora-id1>": 0.6, "<lora-id2>": 0.4}

API Flow

API流程

Submit generation request with
```
X-ModelScope-Async-Mode: true
```
Receive
```
task_id
```
in response

Poll

/v1/tasks/{task_id}

with

X-ModelScope-Task-Type: image_generation

Wait for status
```
SUCCEED
```
or
```
FAILED
```
Download image from
```
output_images[0]
```
URL

提交生成请求时携带
```
X-ModelScope-Async-Mode: true
```
在响应中接收
```
task_id
```

携带

X-ModelScope-Task-Type: image_generation

轮询

/v1/tasks/{task_id}

等待状态变为
```
SUCCEED
```
或
```
FAILED
```
从
```
output_images[0]
```
对应的URL下载图片