FiftyOne
Enables AI assistants to explore computer vision datasets, execute operators, and build workflows through natural language using FiftyOne's operator framework with 80+ built-in operators and plugin management capabilities.
What is FiftyOne?
FiftyOne is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables ai assistants to explore computer vision datasets, execute operators, and build workflows through natural language using fiftyone's operator framework with 80+ built-in operators and plugin ma...
Enables AI assistants to explore computer vision datasets, execute operators, and build workflows through natural language using FiftyOne's operator framework with 80+ built-in operators and plugin management capabilities.
This server falls under the Data Science & ML category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Enables AI assistants to explore computer vision datasets, e
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx fiftyoneConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use FiftyOne
The FiftyOne MCP Server bridges AI assistants and the FiftyOne computer vision dataset management platform, exposing over 45 tools for dataset exploration, sample management, operator execution, and plugin administration through natural language. It allows AI tools like Claude to load datasets, filter samples by confidence or label, launch the FiftyOne App, run aggregations, and install or manage plugins without writing any Python code directly. ML engineers and data scientists use it to accelerate dataset curation, quality review, and workflow automation in computer vision pipelines.
Prerequisites
- Python environment with fiftyone installed (pip install fiftyone)
- fiftyone-mcp-server installed in the same Python environment
- An MCP-compatible client (Claude Desktop, Claude Code, Cursor, VS Code)
- At least one FiftyOne dataset loaded or accessible locally
Install fiftyone and fiftyone-mcp-server
Install both FiftyOne and the MCP server package into the same Python environment so the server can access your datasets.
pip install fiftyone fiftyone-mcp-serverAdd the server to Claude Code (recommended)
Use the claude mcp add command for the simplest setup. This registers the fiftyone-mcp command as an MCP server in your Claude Code configuration.
claude mcp add fiftyone -- fiftyone-mcpOr configure Claude Desktop manually
If using Claude Desktop, add the server configuration to the claude_desktop_config.json file.
Verify the server starts correctly
Run the server manually to confirm it starts without errors before connecting an AI client.
fiftyone-mcpLoad a dataset and begin exploring
With the server connected, ask your AI assistant to list available datasets or load the built-in quickstart dataset to verify everything works.
FiftyOne Examples
Client configuration
Claude Desktop configuration using the installed fiftyone-mcp command, and an alternative using uvx for no-install execution.
{
"mcpServers": {
"fiftyone": {
"command": "fiftyone-mcp"
}
}
}
// Alternative with uvx (no separate install):
{
"mcpServers": {
"fiftyone": {
"command": "uvx",
"args": ["fiftyone-mcp-server"]
}
}
}Prompts to try
Example natural language prompts that use FiftyOne's dataset management and operator tools.
- "List all my FiftyOne datasets"
- "Load the quickstart dataset and show me a summary"
- "Select samples with prediction confidence above 0.9"
- "Find and tag duplicate images in the current dataset"
- "Install the brain plugin"
- "Run the compute_uniqueness operator on the loaded dataset"
- "Show me the label distribution histogram for the detection labels"Troubleshooting FiftyOne
The fiftyone-mcp command is not found after installation
Ensure you are using the same Python environment where fiftyone-mcp-server was installed. If using virtual environments, activate it before running fiftyone-mcp. Check 'which fiftyone-mcp' to confirm the binary location.
Dataset operations fail with 'no dataset loaded' errors
FiftyOne datasets must exist in your local MongoDB instance. First create or import a dataset using Python (e.g., fo.load_dataset('quickstart') or fo.Dataset.from_dir(...)). The MCP server cannot create datasets from raw files directly.
App operations time out or the FiftyOne App does not launch
The launch_app tool starts a local FiftyOne App server. Ensure no other FiftyOne App instance is already running on the default port (5151). Restart Python and the MCP server if a stale session is blocking the port.
Frequently Asked Questions about FiftyOne
What is FiftyOne?
FiftyOne is a Model Context Protocol (MCP) server that enables ai assistants to explore computer vision datasets, execute operators, and build workflows through natural language using fiftyone's operator framework with 80+ built-in operators and plugin management capabilities. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install FiftyOne?
Follow the installation instructions on the FiftyOne GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with FiftyOne?
FiftyOne works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is FiftyOne free to use?
Yes, FiftyOne is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
FiftyOne Alternatives — Similar Data Science & ML Servers
Looking for alternatives to FiftyOne? Here are other popular data science & ml servers you can use with Claude, Cursor, and VS Code.
Ultrarag
★ 5.6kA Low-Code MCP Framework for Building Complex and Innovative RAG Pipelines
RocketRide
★ 3.1k📇 🏠 - MCP server that exposes RocketRide AI pipelines as t
Aix Db
★ 2.1kAix-DB 基于 LangChain/LangGraph 框架，结合 MCP Skills 多智能体协作架构，实现自然语言到数据洞察的端到端转换。
NeMo Data Designer
★ 1.9k🎨 NeMo Data Designer: Generate high-quality synthetic data from scratch or from seed data.
PaperBanana
★ 1.7kOpen source implementation and extension of Google Research’s PaperBanana for automated academic figures, diagrams, and research visuals, expanded to new domains like slide generation.
MiniMax
★ 1.5kBridges MiniMax AI capabilities to the Model Context Protocol, enabling AI agents to perform image understanding, text-to-image generation, and speech synthesis. It provides a standardized interface for accessing MiniMax's core tools via JSON-RPC.
Browse More Data Science & ML MCP Servers
Explore all data science & ml servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up FiftyOne in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use FiftyOne?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.