mizall90/mm-mcp
If you are the rightful owner of mm-mcp and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
This is a ModelManager MCP server that provides a REST API for ModelManager tools.
The ModelManager MCP Server is designed to facilitate the integration and management of machine learning models through a REST API. It serves as a bridge between various ModelManager tools and the underlying infrastructure, allowing for seamless deployment, scaling, and monitoring of models. The server can be configured to run locally or within a Docker container, providing flexibility in deployment options. It supports secure communication through environment variables and can be integrated with different platforms such as windsurf, vscode, and claude desktop.
Features
- REST API for ModelManager tools
- Local and Docker deployment options
- Secure communication with environment variables
- Integration with multiple platforms
- Scalable and flexible model management
Usages
local configuration
{
"mcpServers": {
"mm-mcp": {
"command": "mm-mcp/.venv/bin/mcp",
"args": ["run", "mm-mcp/server/mm.py"]
}
}
}docker configuration
{
"mcpServers": {
"mm-mcp-docker": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"--network=host",
"-e",
"SECRET_KEY",
"-e",
"MM_API_BASE_URL",
"modelmanagerdev/mcp:v1"
],
"env": {
"SECRET_KEY": "your-secret-key",
"MM_API_BASE_URL": "your-api-base-url"
}
}
}
}
