# imagine-mcp by n24q02m

n24q02m’s imagine-mcp provides image and video understanding and generation tools for AI agents across supported model providers.

Canonical URL: https://a2a-mcp.org/entry/imagine-mcp-n24q02m

Language: en

Category: MCP Servers

Updated: 2026-09-14

## About

n24q02m’s imagine-mcp provides image and video understanding and generation tools for AI agents across supported model providers.

## Key Features

- Use multimodal understanding and generation in an agent workflow.
- Connect supported Gemini, OpenAI and Grok capabilities through one project.

## Use Cases

Add model inference or media generation tools to a developer-controlled agent.

## Limitations

This is an independent open-source project. It is not ImagineArt’s hosted MCP connector, which is already listed separately.

## Sources

- Project documentation: https://github.com/n24q02m/imagine-mcp

## MCP Setup

### Prerequisites

- uv/uvx for the published Python package.

### Access requirements

- At least one supported provider API key, such as Gemini, OpenAI or Grok.

### Client setup

#### Other

Transport: `stdio`

```json
{
  "mcpServers": {
    "imagine": {
      "command": "uvx",
      "args": [
        "imagine-mcp"
      ],
      "env": {
        "GEMINI_API_KEY": "YOUR_GEMINI_API_KEY"
      }
    }
  }
}
```

- Replace the example provider key or follow the setup guide for a different provider.

### Environment variables

- `GEMINI_API_KEY (example provider)`

Setup docs: https://github.com/n24q02m/imagine-mcp

## Tags

- MCP
- media-generation
- multimodal

## Also known as

- n24q02m imagine-mcp

## Alternatives

- fal MCP Server
- Replicate MCP Server

## Sources

- [Project website](https://github.com/n24q02m/imagine-mcp)
- [Project documentation](https://github.com/n24q02m/imagine-mcp)

Last checked: 2026-09-14