OpenAI GPT Image MCP Server
MCP server for working with OpenAI GPT-4o/gpt-image-1 image generation and editing APIs, supporting image creation from text prompts and advanced editing capabilities.
Get this MCP server
MCP server for working with OpenAI GPT-4o/gpt-image-1 image generation and editing APIs, supporting image creation from text prompts and advanced editing capabilities.
Installation
From Source Code
git clone https://github.com/SureScaleAI/openai-gpt-image-mcp.git
cd openai-gpt-image-mcp
yarn install
yarn build
Configuration
Claude Desktop/VSCode/Cursor/Windsurf
{
"mcpServers": {
"openai-gpt-image-mcp": {
"command": "node",
"args": ["/absolute/path/to/dist/index.js"],
"env": { "OPENAI_API_KEY": "sk-..." }
}
}
}
Azure OpenAI Configuration
{
"mcpServers": {
"openai-gpt-image-mcp": {
"command": "node",
"args": ["/absolute/path/to/dist/index.js"],
"env": {
"AZURE_OPENAI_API_KEY": "sk-...",
"AZURE_OPENAI_ENDPOINT": "my.endpoint.com",
"OPENAI_API_VERSION": "2024-12-01-preview"
}
}
}
}
With Environment File
{
"mcpServers": {
"openai-gpt-image-mcp": {
"command": "node",
"args": ["/absolute/path/to/dist/index.js", "--env-file", "./deployment/.env"]
}
}
}
Available Tools
| Tool | Description |
|---|---|
create-image |
Generates images from prompts with advanced options (size, quality, background, etc.) |
edit-image |
Edits or extends images using a prompt and optional mask, supports both file paths and base64 |
Features
- Image generation from text prompts using the latest OpenAI models
- Image editing (inpainting, outpainting, compositing) with advanced control through prompts
- Support for Claude Desktop, Cursor, VSCode, Windsurf, and any MCP-compatible clients
- File output: save generated images directly to disk or retrieve as base64
- Generate up to 10 images simultaneously using create-image
- Mask support for controlling editing areas
- Automatic file output switching for large images exceeding the 1MB limit
Environment Variables
Required
OPENAI_API_KEY- OpenAI API key for accessing the image generation API
Optional
AZURE_OPENAI_API_KEY- Azure OpenAI API key for Azure deploymentsAZURE_OPENAI_ENDPOINT- Azure OpenAI endpoint URLOPENAI_API_VERSION- OpenAI API version for Azure deploymentsMCP_HF_WORK_DIR- Directory for saving large images and file outputs
Notes
A verified OpenAI organization is required to access the image API. After verification, activating image API access may take 15-20 minutes. MCP clients have a 1MB limit, so large images are automatically saved to disk. File paths must be absolute. Configuration via environment file with the --env-file flag is supported.