# TopazLabs/topaz-mcp [Health: Active]

**Category:** 🎥 Multimedia Process  
**Repository:** https://github.com/TopazLabs/topaz-mcp  
**GitHub Stars:** 6  
**npm Downloads (last month):** 146  
**Views:** 4  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/topazlabs-topaz-mcp

## Description
AI image enhancement (upscaling, denoising, sharpening) via Topaz Labs API. Supports 8 models including Standard V2, Wonder 2, Bloom, and Recover 3.

## Tools
Capabilities this server exposes over MCP:

- **image-enhance** — Enhance an image using Topaz Labs AI. Upscales, sharpens, and improves quality. Accepts a local file path or URL. If the user shares an image in the conversation without providing a path, ask them for the file path or a URL to the original image.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "topaz-mcp": {
    "command": "npx",
    "args": ["-y","@topazlabs/mcp"],
    "env": {
      "TOPAZ_API_KEY": ""
    }
  }
}
```

**Requires environment variables:** `TOPAZ_API_KEY` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What TopazLabs/topaz-mcp MCP server does

TopazLabs/topaz-mcp MCP server exposes Topaz Labs image enhancement through the Model Context Protocol. An assistant can submit a local image path or URL and receive an enhanced image. The available processing is image-only and runs through Topaz Labs' cloud API, so the source image is sent to that service.

The server supports eight models: Standard V2, Low Resolution V2, CGI, High Fidelity V2, Redefine, Recover 3, Standard MAX, and Wonder 2. These cover general photography, small or low-resolution images, digital art, high-quality source images, creative reinterpretation, old photographs, maximum-quality processing, and nature or portrait detail.

## How it works

The MCP server uses stdio transport and registers the `image-enhance` tool. The required `image` input can be a filesystem path or URL. A request may also include a `model` and a `size` object. Size can specify exact `width` and `height`, only one dimension for proportional resizing, or a `scale` multiplier.

If no model is chosen, processing uses Standard V2. If no size is provided, images smaller than 2K are automatically upscaled to 2K width. Images shared directly in a chat are not sufficient by themselves; the assistant needs the original file path or a URL. Clients with filesystem access are therefore better suited to local-image workflows.

## Setup and configuration

To run the TopazLabs/topaz-mcp MCP server, install it through npm when configuring an MCP client:

```json
{
  "mcpServers": {
    "Topaz Labs": {
      "command": "npx",
      "args": ["-y", "@topazlabs/mcp"],
      "env": {
        "TOPAZ_API_KEY": "your-api-key"
      }
    }
  }
}
```

`TOPAZ_API_KEY` is required. The key is obtained from the Topaz Labs developer portal. The documented configuration uses stdio, which is supported by Claude Desktop, Cursor, Windsurf, and Cline. Claude Desktop's chat mode requires the user to provide paths manually because it does not provide filesystem access; the README recommends Claude Cowork for local files.

## Tools and capabilities

The server exposes one tool:

- `image-enhance`: Enhances an image using a Topaz Labs model. It can upscale, sharpen, and improve image quality from a local path or URL.

The tool's model choices include standard and generative options. Standard V2 is the default for general photos, while Low Resolution V2 targets tiny images and thumbnails. CGI is intended for digital art and renders; High Fidelity V2 is for high-quality sources requiring enlargement. Redefine, Recover 3, Standard MAX, and Wonder 2 provide generative processing for creative reinterpretation, older or low-resolution images, maximum quality, and nature photography or portraits respectively.

## Limitations and notes

The TopazLabs/topaz-mcp MCP server requires a paid Topaz Labs API key and sends images to the Topaz Labs API for cloud processing. It does not enhance video. HTTP streaming transport is not currently supported, so clients that require a remote HTTP or web/mobile connection cannot use that transport with this server.

_Full upstream README: https://allmcps.com/mcp/topazlabs-topaz-mcp/readme_