MCP Server for Ollama
@vincentf305
About MCP Server for Ollama
MCP server for connecting Claude Desktop to Ollama LLM server
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"ollama-server": {
"command": "python",
"args": [
"-m",
"src.mcp_server.server"
],
"env": {
"PYTHONPATH": "path-to-mcp-server"
}
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is MCP Server for Ollama?
MCP Server for Ollama is a Model Control Protocol (MCP) server that enables Claude Desktop to communicate with an Ollama LLM server. It is intended for users who want to run local large language models via Ollama and interact with them through Claude Desktop.
How to use MCP Server for Ollama?
Clone the repository, copy .env.example to .env and configure as needed, then install dependencies with pip install -r requirements.txt. To use with Claude Desktop, edit claude_desktop_config.json to include the server definition, setting the command to python -m src.mcp_server.server and the PYTHONPATH environment variable to the repo path.
Key features of MCP Server for Ollama
- Integrates Claude Desktop with Ollama via Model Control Protocol
- Simple setup with pip and a configuration file
- Allows use of local Ollama models from Claude Desktop
- Environment variable configuration via
.env
Use cases of MCP Server for Ollama
- Query local Ollama models directly from Claude Desktop
- Run privacy-preserving LLM interactions without sending data to external APIs
- Develop and test with different Ollama models through Claude's interface
FAQ from MCP Server for Ollama
What does MCP Server for Ollama do?
It provides a Model Control Protocol server that lets Claude Desktop communicate with an Ollama LLM server for inference.
How do I set it up with Claude Desktop?
Clone the repo, set up .env, install Python dependencies, then add the server entry to claude_desktop_config.json as shown in the README.
What are the dependencies?
Python 3, the packages listed in requirements.txt (install via pip), and access to a running Ollama server.
Does it require Ollama to be installed?
Yes, it communicates with an Ollama LLM server, so Ollama must be running and accessible.
Where are the model weights and data stored?
The MCP server itself doesn’t store them; data is processed by the local Ollama instance according to its own configuration.
Frequently asked questions
What does MCP Server for Ollama do?
It provides a Model Control Protocol server that lets Claude Desktop communicate with an Ollama LLM server for inference.
How do I set it up with Claude Desktop?
Clone the repo, set up `.env`, install Python dependencies, then add the server entry to `claude_desktop_config.json` as shown in the README.
What are the dependencies?
Python 3, the packages listed in `requirements.txt` (install via pip), and access to a running Ollama server.
Does it require Ollama to be installed?
Yes, it communicates with an Ollama LLM server, so Ollama must be running and accessible.
Where are the model weights and data stored?
The MCP server itself doesn’t store them; data is processed by the local Ollama instance according to its own configuration.
Basic information
More AI & Agents MCP servers
21st.dev Magic AI Agent
21st-devIt's like v0 but in your Cursor/WindSurf/Cline. 21st dev Magic MCP server for working with your frontend like Magic
fhirHydrant
faulkjOpen-source Node.js FHIR MCP server with SMART Backend Services, metadata-aware search/CRUD tools, compact responses, FHIRPath filtering, safe pagination, audit events, and terminology lookup.
Heterogent
Hire AI agents and lock USDC in trustless on-chain escrow on Solana all from your AI assistant. Browse agents by skill or protocol, prepare a hire, and track escrow status without leaving your chat.
Hypnothera
La-SalidaAgent skill: turn what your AI assistant knows about you into a personalized hypnosis session on hypnothera.ai
Agentic Atlas
Avery JonesConsult field-tested agent-system design patterns through 8 hosted, read-only tools for orientation, deep reading, relationships, provenance, glossary terms, and publication decisions.
Comments