Use fast-agent to use MCP tools with local LLM, API or Claude Desktop. WIP
Overview
What is Ollama MCP Server?
Ollama MCP Server is a Model Context Protocol (MCP) server that allows Claude to run Ollama models asynchronously, storing outputs for later retrieval. It is built with uv for Python environment management.
How to use Ollama MCP Server?
To use the Ollama MCP Server, integrate it with Claude Desktop by copying the configuration from claude_desktop_config.json and adjusting file paths as necessary. You can then run models, execute scripts, and manage jobs through the provided tools.
Key features of Ollama MCP Server?
- Asynchronous execution of Ollama models
- Management of script templates with variable substitution
- Execution of bash commands and multi-step workflows
- Storage of all outputs in a dedicated directory
- Simple configuration for integration with Claude Desktop
Use cases of Ollama MCP Server?
- Running complex models without waiting for completion.
- Automating workflows that involve multiple steps and commands.
- Managing and executing script templates for various tasks.
FAQ from Ollama MCP Server?
- Can I run multiple models at the same time?
Yes! The server supports asynchronous execution, allowing multiple models to run concurrently.
- How do I check the status of a job?
You can use the
get_job_statustool to check if a job is completed or still running.
- Is there a limit to the number of scripts I can create?
No, you can create as many script templates as needed.