Py GPT
Desktop AI Assistant powered by GPT-5, GPT-4, o1, o3, Gemini, Claude, Ollama, DeepSeek, Perplexity, Grok, Bielik, chat, vision, voice, RAG, image and video generation, agents, tools, MCP, plugins, speech synthesis and recognition, web search, memory,
What is Py GPT?
Py GPT is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to desktop ai assistant powered by gpt-5, gpt-4, o1, o3, gemini, claude, ollama, deepseek, perplexity, grok, bielik, chat, vision, voice, rag, image and video generation, agents, tools, mcp, plugins, spe...
Desktop AI Assistant powered by GPT-5, GPT-4, o1, o3, Gemini, Claude, Ollama, DeepSeek, Perplexity, Grok, Bielik, chat, vision, voice, RAG, image and video generation, agents, tools, MCP, plugins, speech synthesis and recognition, web search, memory,
This server falls under the Developer Tools category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Desktop AI Assistant powered by GPT-5, GPT-4, o1, o3, Gemini
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx py-gptConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Py GPT
PyGPT is a cross-platform desktop AI assistant that integrates GPT-5, GPT-4, o1, o3, Claude, Gemini, DeepSeek, and local Ollama models into a single application with 11 operational modes including chat, vision, voice, image generation, code interpretation, and autonomous agent operation. It includes native MCP support via a plugin that connects to any stdio, SSE, or Streamable HTTP MCP server with per-server allow/deny filtering, making it a full MCP Host for desktop users. Developers and researchers use PyGPT when they want a self-contained desktop environment for experimenting with multiple AI models alongside local file operations, plugins, and RAG workflows.
Prerequisites
- Python 3.10+ installed on Windows, macOS, or Linux
- pip package manager or the Snap package manager on Linux
- At least one provider API key: OPENAI_API_KEY, ANTHROPIC_API_KEY, GOOGLE_API_KEY, or Ollama running locally
- Optional: HUGGING_FACE_TOKEN for HuggingFace model access
- Optional: audio input/output device for voice mode
Install PyGPT
Install PyGPT from PyPI using pip. On Linux you can also install via Snap. The main package is pygpt-net.
pip install pygpt-net
# Linux Snap:
sudo snap install pygpt
# Then launch:
pygptConfigure API keys through the GUI
Launch PyGPT and open Config → Settings → API Keys. Enter your provider credentials (OpenAI, Anthropic, Google, etc.). Settings are saved to config.json in the working directory.
Select a model and operational mode
In the main interface, choose a model from the dropdown and select an operational mode: Chat, Chat with Files (RAG), Image Generation, Code Interpreter, Agent, Vision, Voice, or Web Search. Each mode adjusts which tools and plugins are active.
Enable the MCP plugin
Go to Plugins → MCP and enable it. Add your MCP server connections by entering the server name, transport type (stdio, SSE, or Streamable HTTP), and command or URL. Configure per-server allow/deny lists for tools.
Configure MCP server connection in the plugin
For stdio-based MCP servers, enter the command and arguments. For SSE or HTTP servers, enter the endpoint URL. PyGPT will start stdio servers automatically and maintain persistent connections to HTTP servers.
Set up Ollama for local inference (optional)
If you want fully offline operation, install Ollama and pull a model. Set OLLAMA_API_BASE to point PyGPT at your local Ollama instance, then select an Ollama model in the model dropdown.
# Set environment variable before launching PyGPT:
export OLLAMA_API_BASE=http://localhost:11434
pygptPy GPT Examples
MCP plugin configuration example
PyGPT's MCP plugin configuration is managed through the GUI, but the underlying config.json stores server definitions in this format.
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/home/user/documents"],
"transport": "stdio"
},
"fetch": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-fetch"],
"transport": "stdio"
}
}
}Prompts to try
Once PyGPT is running with your chosen model and MCP plugin enabled, try these prompts in the Chat or Agent mode.
- "Read the file /home/user/notes.txt and summarize its key points"
- "Search the web for recent AI news and create a briefing document"
- "Generate an image of a futuristic city skyline and save it to my desktop"
- "Analyze the code in my project folder and identify potential bugs"
- "Transcribe my voice message and translate it to French"Troubleshooting Py GPT
PyGPT fails to start with a missing dependency error
Install from source to ensure all dependencies are resolved: git clone the repo, create a virtual environment (python3 -m venv venv && source venv/bin/activate), then pip install -r requirements.txt && python3 run.py.
MCP plugin cannot connect to a stdio server
Verify the command is on PATH by running it manually in a terminal. Ensure Node.js is installed for npx-based servers. Check the plugin's per-server allow list to confirm the desired tools are not blocked.
Ollama models are not listed in PyGPT
Ensure Ollama is running (ollama serve) and OLLAMA_API_BASE is set to http://localhost:11434. Pull the desired model first with ollama pull <model-name> before launching PyGPT.
Frequently Asked Questions about Py GPT
What is Py GPT?
Py GPT is a Model Context Protocol (MCP) server that desktop ai assistant powered by gpt-5, gpt-4, o1, o3, gemini, claude, ollama, deepseek, perplexity, grok, bielik, chat, vision, voice, rag, image and video generation, agents, tools, mcp, plugins, speech synthesis and recognition, web search, memory, It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Py GPT?
Follow the installation instructions on the Py GPT GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Py GPT?
Py GPT works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Py GPT free to use?
Yes, Py GPT is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.
Py GPT Alternatives — Similar Developer Tools Servers
Looking for alternatives to Py GPT? Here are other popular developer tools servers you can use with Claude, Cursor, and VS Code.
Ecc
★ 188.2kThe agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Javaguide
★ 155.8kJava 面试 & 后端通用面试指南，覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Gemini CLI
★ 104.5kA secure MCP server that wraps the Google Gemini CLI, allowing clients to query Gemini models using local OAuth sessions without requiring an API key. It provides tools for model interaction and diagnostics with built-in protection against command in
Awesome MCP Servers
★ 87.3k⭐ Curated list of Model Context Protocol (MCP) servers - tools that extend Claude Desktop, Cursor, Windsurf, and other MCP clients with custom capabilities.
MCP Servers
★ 86.0kModel Context Protocol Servers
CC Switch
★ 77.5kA cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Gemini CLI & Hermes Agent. Only official website: ccswitch.io
Browse More Developer Tools MCP Servers
Explore all developer tools servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Py GPT 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 Py GPT?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.