- Home
- MCP servers
- Novita
Novita
- javascript
12
GitHub Stars
javascript
Language
6 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"novitalabs-novita-mcp-server": {
"command": "npx",
"args": [
"-y",
"@novitalabs/novita-mcp-server"
],
"env": {
"NOVITA_API_KEY": "your_api_key_here"
}
}
}
}You can manage Novita AI GPU resources through a dedicated MCP server that connects your MCP client to the Novita platform. This server lets you list, create, start, stop, and manage GPU instances, templates, clusters, networks, and more from your MCP-enabled client. It’s designed to streamline resource control across your cloud regions and product offerings.
How to use
Use your MCP client (such as Claude Desktop or Cursor) to connect to the novita MCP server you configured. With this server, you can list available clusters, products, and GPU instances, and perform lifecycle actions on GPU instances (create, start, stop, restart, delete). You can also manage templates, container registry authentication entries, and network storage.
How to install
Prerequisites: ensure you have Node.js and npm installed on your system.
Install the MCP server using npm:
npm install -g @novitalabs/novita-mcp-server
Alternatively, install using Smithery by following the installation instructions for the novita-mcp-server on Smithery.
# Smithery command would be used here if you are following that path
Configure the MCP server in your client by providing your Novita API key. The configuration below demonstrates how to run the MCP server via npx and pass your API key as an environment variable.
{
"mcpServers": {
"novita_mcp": {
"command": "npx",
"args": ["-y", "@novitalabs/novita-mcp-server"],
"env": {
"NOVITA_API_KEY": "your_api_key_here"
}
}
}
}
Additional notes and tips
Updated on beta: current release supports GPU instance management and related resources. Other resource types will be added in future releases.
Tools and capabilities
This MCP server exposes a set of operations across several resource types. Typical capabilities include the following functions to manage GPU resources and related infrastructure:
Common operations you can perform
- Cluster/Region: List
- Product: List
- GPU Instance: List, Get, Create, Start, Stop, Delete, Restart
- Template: List, Get, Create, Delete
- Container Registry Auth: List, Create, Delete
- Network Storage: List, Create, Update, Delete
Available tools
Cluster List
List all available clusters (regions) for your Novita resources.
Product List
List all available GPU instance products.
GPU Instance List
List all GPU instances under your account.
GPU Instance Get
Retrieve details for a specific GPU instance.
GPU Instance Create
Create a new GPU instance with specified name, product, GPU count, image, and disk size.
GPU Instance Start
Start a stopped GPU instance.
GPU Instance Stop
Stop a running GPU instance.
GPU Instance Delete
Delete a GPU instance.
GPU Instance Restart
Restart a GPU instance.
Template List
List available templates for GPU configurations.
Template Get
Get details of a specific template.
Template Create
Create a new resource template.
Template Delete
Delete a template.
Container Registry Auth List
List container registry authentication entries.
Container Registry Auth Create
Create a new container registry authentication entry.
Container Registry Auth Delete
Delete a container registry authentication entry.
Network Storage List
List network storage resources.
Network Storage Create
Create a new network storage resource.
Network Storage Update
Update an existing network storage resource.
Network Storage Delete
Delete a network storage resource.