O
Ollama
Run local Ollama models over MCP.
·No reviews yet
Share:
About
Auto-generated
Ollama is a Model Context Protocol server in the Developer Tools category. It lets AI assistants that speak MCP call its tools and read its resources over STDIO.
Example tools
Illustrative tool names — the actual tool set is defined by the server at runtime.
list_resources
Enumerate resources exposed by this MCP
get_resource
Fetch a resource by ID
invoke
Invoke the primary action of this MCP
Screenshots
No screenshots yet.
Frequently Asked Questions
Common questions about Ollama.
- What is the Ollama MCP for?
- The Ollama MCP allows users to run large language models (LLMs) served by a local Ollama instance through applications that support the Model Context Protocol. It enables local, offline LLM interactions for development and testing.
- How do I install the Ollama MCP for use with a client like Cursor?
- To use the Ollama MCP with a client, you typically add it to your client's MCP configuration by specifying its stdio transport and ollama slug. The exact steps depend on your client's interface for adding custom MCPs.
- Does the Ollama MCP require an API key or other authentication?
- No, the Ollama MCP does not require an API key, OAuth, or any other authentication credentials. It is designed to run locally with an existing Ollama installation.
- What types of capabilities does the Ollama MCP provide?
- The Ollama MCP facilitates running local LLM models using the Ollama service. It exposes the ability to interact with these models for tasks such as text generation, summarization, and local development workflows.
- Are there any costs associated with using the Ollama MCP?
- There are no direct costs for the Ollama MCP itself. However, it requires a local Ollama installation, and the models you choose to run with Ollama may have their own licensing terms or resource requirements.
- What is a prerequisite for running the Ollama MCP?
- A key prerequisite for using the Ollama MCP is having Ollama installed and running locally on your machine. This MCP acts as an interface to your existing local Ollama setup.
Install Ollama
Claude Desktop
Add this to claude_desktop_config.json.
{
"mcpServers": {
"ollama": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-ollama"
]
}
}
}Cursor
Add this to ~/.cursor/mcp.json.
{
"mcpServers": {
"ollama": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-ollama"
]
}
}
}VS Code
Add this to your workspace settings.json.
{
"mcp.servers": {
"ollama": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-ollama"
]
}
}
}Reviews
Sign in to leave a review.
No reviews yet. Be the first!
Discussion(0)
No comments yet.