Marimo Notebook

Marimo is a modern, reactive Python notebook — a popular replacement for Jupyter. Cells re-execute automatically when dependencies change, notebooks are stored as clean .py files (not JSON), and there’s no hidden state from out-of-order execution. Carolina Cloud offers two Marimo flavors:

• Marimo — A clean notebook environment. Install what you need.
• Data Science Notebook — Marimo pre-loaded with polars, pandas, scikit-learn, PyTorch, matplotlib, plotly, and 20+ other data science packages. GPU-accelerated libraries (RAPIDS, cuDF) included when you add a GPU.

Via the Dashboard

1. Sign in at console.carolinacloud.io
2. In the sidebar, click Create then General (for basic Marimo) or Data Science (for the full data science stack)
3. Click the Marimo Notebook or Data Science Notebook card
4. Select your CPU tier and optionally add a GPU
5. Use the sliders to configure vCPUs, RAM, and disk size
6. Optionally expand S3 Integration to pre-configure aws s3 access on the instance. You can click “Use my Carolina Cloud bucket” to auto-fill credentials if you have Carolina Cloud storage set up.
7. Review the cost and click Create
8. Once the status changes to “Running”, click the Open button on your instance card to launch the notebook in your browser

Via the API

# Basic Marimo notebook
curl -X POST https://console.carolinacloud.io/api/instance/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "resource_type": "container",
    "flavor": "marimo",
    "n_vcpus": 4,
    "mem_gib": 8,
    "disk_size_gib": 50,
    "cpu_tier": "general-purpose"
  }'

# Data Science notebook with GPU
curl -X POST https://console.carolinacloud.io/api/instance/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "resource_type": "container",
    "flavor": "datasciencemarimo",
    "n_vcpus": 16,
    "mem_gib": 64,
    "disk_size_gib": 200,
    "cpu_tier": "general-purpose",
    "gpu_model": "RTX 5090"
  }'

The response includes a webapp_url field — that’s the URL to open your notebook.

Optional fields:

Field	Description
name	Human-friendly name
cpu_tier	general-purpose (default) or high-performance
gpu_model	GPU model name (e.g. RTX 5090), omit for CPU-only
aws_access_key_id	S3 access key for pre-configured aws s3 access
aws_secret_access_key	S3 secret key
aws_endpoint_url	S3 endpoint (required for non-AWS providers)
aws_prefill_bucket	Bucket path to pre-download into ~/s3-prefill/ on startup

Via the CLI

# Basic Marimo notebook
ccloud new container --cpus 4 --ram 8 --disk 50 --flavor marimo

# Data Science notebook with name
ccloud new container --cpus 16 --ram 64 --disk 200 \
  --flavor datasciencemarimo \
  --name my-analysis

After creation, the notebook URL will be displayed. You can also find it in ccloud list or on the dashboard.

Accessing your notebook

Marimo notebooks are accessed through your web browser. Once the instance is running:

• Click the Open button on the instance card in the dashboard, or
• Navigate directly to https://<instance-name>.carolinacloud.io

When prompted for a password, use the password shown on your instance card (labeled “Gateway Token” for some flavors).

Learn more

For detailed Marimo documentation, see the Marimo User Guide.