ComfyUI OpenAI Image API

A ComfyUI node that provides access to OpenAI's image generation and editing capabilities, including support for gpt-image-1 model with both OpenAI and Azure OpenAI providers.

📋 Project Origin

This project is based on the original work from unicough/comfy_openai_image_api.

Key Enhancements:

• ✅ Added Azure OpenAI support
• ✅ Added flexible configuration options (environment variables + node parameters)
• ✅ Enhanced error handling and security
• ✅ Comprehensive documentation and examples

Features

• Image Generation: Create images from text prompts using gpt-image-1

• Image Editing: Edit existing images with text prompts

• Multiple Providers: Support for both OpenAI and Azure OpenAI

• Quality Control: Low, medium, and high quality options

• Size Options: 1024x1024, 1536x1024, 1024x1536

• Batch Processing: Handle multiple images at once

• Environment Variables: Secure credential management

• Prompt only with no input image:

• One image input:

• Multiple images input:

[!NOTE] This projected was created with a cookiecutter template. It helps you start writing custom nodes without worrying about the Python setup.

Installation

1. Clone this repository into your ComfyUI custom nodes directory:

cd ComfyUI/custom_nodes
git clone https://github.com/your-username/comfy_openai_image_api.git

2. Install the required dependencies:

cd comfy_openai_image_api
pip install -r requirements.txt

3. Restart ComfyUI

Configuration

Environment Variables (Recommended)

Create a .env file in the project root (copy from .env.example):

# OpenAI Configuration
OPENAI_API_KEY=your_openai_api_key_here

# Azure OpenAI Configuration
AZURE_OPENAI_ENDPOINT=https://your-resource-name.openai.azure.com/
AZURE_OPENAI_API_KEY=your_azure_openai_api_key_here
AZURE_OPENAI_API_VERSION=2024-12-01-preview
AZURE_OPENAI_DEPLOYMENT=gpt-image-1

Node Parameters

The node accepts the following parameters:

Required Parameters:

• prompt: Text description of the image you want to generate or edit
• model: Currently supports "gpt-image-1"
• size: Image dimensions (1024x1024, 1536x1024, 1024x1536)
• quality: Image quality (low, medium, high)
• provider: Choose between "openai" or "azure"

Optional Parameters:

• image: Input image for editing (optional, for generation leave empty)
• api_key: API key (can be provided here or via environment variable)
• azure_endpoint: Azure OpenAI endpoint URL (for Azure provider)
• azure_api_version: Azure OpenAI API version (default: 2024-12-01-preview)
• azure_deployment: Azure OpenAI deployment name (default: gpt-image-1)

Usage

OpenAI Provider

1. Set your OpenAI API key in the environment variable OPENAI_API_KEY or provide it in the node
2. Select "openai" as the provider
3. Configure your prompt and other parameters
4. Run the node

Azure OpenAI Provider

1. Set up your Azure OpenAI resource and deploy the gpt-image-1 model
2. Configure environment variables or provide parameters directly:
   • AZURE_OPENAI_ENDPOINT
   • AZURE_OPENAI_API_KEY
   • AZURE_OPENAI_DEPLOYMENT
3. Select "azure" as the provider
4. Configure your prompt and other parameters
5. Run the node

Image Generation

Connect the node without any input image to generate new images from text prompts.

Image Editing

Connect an existing image to the image input to edit/modify the image based on your text prompt.

Examples

Basic Image Generation

Provider: openai
Prompt: "A beautiful landscape with mountains and lakes"
Model: gpt-image-1
Size: 1024x1024
Quality: high

Image Editing with Azure OpenAI

Provider: azure
Prompt: "Add a sunset sky to this image"
Model: gpt-image-1
Size: 1024x1024
Quality: medium
Image: [Connected input image]
Azure Endpoint: https://your-resource.openai.azure.com/
Azure Deployment: gpt-image-1

Error Handling

The node includes comprehensive error handling for:

• Missing or invalid API keys
• Network connectivity issues
• Invalid image formats
• API rate limiting
• Service unavailability

All errors are displayed in the ComfyUI console with detailed messages.

Security Best Practices

• Use environment variables for API keys
• Never commit API keys to version control
• Use Azure Managed Identity when possible
• Regularly rotate API keys
• Monitor API usage and costs

Requirements

• Python 3.8+
• ComfyUI
• OpenAI Python library
• torch
• PIL (Pillow)
• numpy

Quickstart

1. Install ComfyUI.
2. Install ComfyUI-Manager
3. Look up this extension in ComfyUI-Manager. If you are installing manually, clone this repository under ComfyUI/custom_nodes.
4. Restart ComfyUI.

Features

• A list of features

Develop

To install the dev dependencies and pre-commit (will run the ruff hook), do:

cd openai_image_api
pip install -e .[dev]
pre-commit install

The -e flag above will result in a "live" install, in the sense that any changes you make to your node extension will automatically be picked up the next time you run ComfyUI.

Publish to Github

Install Github Desktop or follow these instructions for ssh.

1. Create a Github repository that matches the directory name.
2. Push the files to Git

git add .
git commit -m "project scaffolding"
git push

Writing custom nodes

An example custom node is located in node.py. To learn more, read the docs.

Tests

This repo contains unit tests written in Pytest in the tests/ directory. It is recommended to unit test your custom node.

• build-pipeline.yml will run pytest and linter on any open PRs
• validate.yml will run node-diff to check for breaking changes

Publishing to Registry

If you wish to share this custom node with others in the community, you can publish it to the registry. We've already auto-populated some fields in pyproject.toml under tool.comfy, but please double-check that they are correct.

You need to make an account on https://registry.comfy.org and create an API key token.

• [ ] Go to the registry. Login and create a publisher id (everything after the @ sign on your registry profile).
• [ ] Add the publisher id into the pyproject.toml file.
• [ ] Create an api key on the Registry for publishing from Github. Instructions.
• [ ] Add it to your Github Repository Secrets as REGISTRY_ACCESS_TOKEN.

A Github action will run on every git push. You can also run the Github action manually. Full instructions here. Join our discord if you have any questions!

🚀 Azure OpenAI 图像编辑功能

新增功能

✅ 增强的 Azure OpenAI 集成

• 使用最新的 Azure OpenAI SDK (2025-04-01-preview)
• 改进的错误处理和日志记录
• 模块化的配置管理
• 类型安全的代码实现

✅ 高级图像处理

• 智能图像格式转换
• 批量图像处理支持
• 图像尺寸验证和调整
• 优化的内存使用

✅ 企业级功能

• 全面的错误处理和重试机制
• 详细的日志记录和监控
• 安全的凭证管理
• 配置验证和健康检查

环境变量配置

项目支持多种环境变量格式，按优先级排序：

# Azure OpenAI 配置
AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com
AZURE_OPENAI_API_KEY=your-api-key
AZURE_OPENAI_API_VERSION=2025-04-01-preview  # 可选
AZURE_OPENAI_DEPLOYMENT=gpt-image-1         # 可选

# 或者使用简化格式
AZURE_ENDPOINT=https://your-resource.openai.azure.com
AZURE_API_KEY=your-api-key

# OpenAI 配置
OPENAI_API_KEY=your-openai-api-key

使用示例

基本图像编辑

from src.openai_image_api.nodes import OpenAIImageAPI
import torch

# 创建节点实例
node = OpenAIImageAPI()

# 准备输入图像 (假设是 ComfyUI 张量格式)
# image_tensor = your_image_tensor

# 进行图像编辑
result = node.generate_image(
    prompt="make it in the style of Studio Ghibli",
    model="gpt-image-1",
    size="1024x1024",
    quality="high",
    provider="azure",
    image=image_tensor,
    azure_endpoint="https://your-resource.openai.azure.com",
    azure_api_version="2025-04-01-preview",
    azure_deployment="gpt-image-1"
)

edited_image = result[0]

使用环境变量

# 设置环境变量后，无需提供额外参数
result = node.generate_image(
    prompt="convert to a cyberpunk style with neon lights",
    model="gpt-image-1", 
    size="1024x1024",
    quality="high",
    provider="azure",
    image=image_tensor
)

运行示例脚本

# 设置环境变量
export AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com
export AZURE_OPENAI_API_KEY=your-api-key

# 运行示例
python examples/azure_image_edit_example.py

配置管理

项目使用模块化配置管理：

from src.openai_image_api.azure_config import AzureConfigManager

# 创建配置
config = AzureConfigManager.create_config(
    endpoint="https://your-resource.openai.azure.com",
    api_key="your-api-key",
    api_version="2025-04-01-preview",
    deployment="gpt-image-1"
)

# 验证配置
AzureConfigManager.validate_config(config)

# 获取配置摘要（隐藏敏感信息）
summary = AzureConfigManager.get_config_summary(config)
print(summary)

图像处理工具

内置的图像处理工具：

from src.openai_image_api.image_utils import ImageProcessor
from PIL import Image
import torch

# PIL 图像转张量
pil_image = Image.open("input.jpg")
tensor = ImageProcessor.pil_to_tensor(pil_image)

# 张量转 PIL 图像
pil_image = ImageProcessor.tensor_to_pil(tensor.squeeze(0))

# 为 API 准备图像数据
images = ImageProcessor.prepare_images_for_api(tensor.squeeze(0))

# Base64 转张量
tensor = ImageProcessor.base64_to_tensor(base64_string)