DINO-X-MCP

Overview

DINO-X MCP is an official server from IDEA-Research that brings fine-grained object detection and image understanding to multimodal applications. It supports multiple transport modes including STDIO and Streamable HTTP, with features like full-scene object detection, text-prompted object detection, human pose estimation, and visualization capabilities. The server provides structured outputs with object categories, counts, locations, and attributes for VQA and multi-step reasoning tasks.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• detect-all-objects

  Full-scene object detection to identify all objects in an image

  Try:Detect all objects in this imageTry:Find all items in the scene
• detect-objects-by-text

  Text-prompted object detection to find specific objects based on text input

  Try:Find all cars in this imageTry:Detect only red objects in the scene
• detect-human-pose-keypoints

  Human pose estimation to detect 17 keypoints in human figures

  Try:Detect human pose in this imageTry:Find all people and their body poses
• visualize-detection-result

  Create visualization of detection results with bounding boxes and labels

  Try:Visualize the detection resultsTry:Draw bounding boxes around detected objects

Comparable tools

Installation

{
  "mcpServers": {
    "dinox-mcp": {
      "url": "https://mcp.deepdataspace.com/mcp?key=your-api-key"
    }
  }
}

{
  "mcpServers": {
    "dinox-mcp": {
      "command": "npx",
      "args": ["-y", "@deepdataspace/dinox-mcp"],
      "env": {
        "DINOX_API_KEY": "your-api-key-here",
        "IMAGE_STORAGE_DIRECTORY": "/path/to/your/image/directory"
      }
    }
  }
}

FAQ

What transport modes are supported?

DINO-X MCP supports STDIO (default) and Streamable HTTP modes. STDIO supports both file:// and https:// image sources, while HTTP supports https:// only.

What image formats are supported?

The server supports jpg, jpeg, webp, and png image formats.

Compare DINO-X-MCP with