Perplexity MCP Server logo

Perplexity MCP Server

by Alcova-AI

The Perplexity MCP Server is a Go-based server that acts as a bridge between AI assistants and the Perplexity API. It allows AI assistants like Claude and Cursor to access Perplexity's search and reasoning capabilities.

View on GitHub

Last updated: N/A

What is Perplexity MCP Server?

The Perplexity MCP Server is a Model Context Protocol (MCP) server that enables AI assistants to seamlessly access Perplexity's search and reasoning capabilities. It acts as a bridge between AI assistants and the Perplexity API.

How to use Perplexity MCP Server?

The server can be installed using Homebrew, from source, or via pre-built binaries. Once installed, it can be configured with AI assistants like Claude Code and Claude Desktop by adding the server's configuration details, including the command to run the server and the Perplexity API key.

Key features of Perplexity MCP Server

  • Access to real-time information

  • Enhanced reasoning capabilities

  • Seamless integration with Claude Code, Claude Desktop, and Cursor

  • Simple installation

  • Customizable model selection

Use cases of Perplexity MCP Server

  • Enabling AI assistants to search the web for up-to-date information

  • Allowing AI assistants to perform complex reasoning tasks

  • Integrating Perplexity's capabilities into AI assistant workflows

  • Providing AI assistants with access to specialized Perplexity models

FAQ from Perplexity MCP Server

What is MCP?

MCP stands for Model Context Protocol, a standard for communication between AI assistants and external tools.

What Perplexity models are supported?

The server supports Perplexity's Sonar Pro model for search and Sonar Reasoning Pro model for reasoning.

How do I get a Perplexity API key?

You need to sign up for a Perplexity API account to obtain an API key.

Can I customize the models used for search and reasoning?

Yes, you can specify the models to use via command-line options or environment variables.

Is direct execution recommended?

Direct execution is not recommended for most users. It's better to use the server through an AI assistant integration.