Interactive Jupyter
Spawn a personal JupyterLab session with the cufemlab client SDK pre-installed. Notebooks persist between sessions in a per-user volume.
Launch a Jupyter session
A fresh container will spawn (~10–15 seconds). JupyterLab opens in a new browser tab for the best experience (full screen, no iframe sandbox quirks, native WebSockets). You can close that tab and resume — your work persists in the per-user volume. The session auto-stops when your daily quota is reached.
Cannot start: your daily quota for tier is used up. Resets at UTC midnight.
Your Jupyter session is running
Container
· cufemlab-jupyter:0.1 with client SDK + viz stack pre-loaded.
Use the inline Code Runner below, or launch the full JupyterLab UI in a new tab.
✓ Link copied to clipboard at — paste into any browser tab.
Notebook
Add cells, type Python, run individually. Shift+Enter = run & go to next · Ctrl+Enter = run & stay · Kernel state persists between cells.
Tier-based quotas
| Tier | Hours / day | GPU access | Resource limits |
|---|---|---|---|
| 1 hour | CPU only | 2 cores · 4 GB RAM | |
| 10 hours | NVIDIA GPU (when available) | 2 cores · 4 GB RAM | |
| 24 hours (effectively unlimited) | NVIDIA GPU (when available) | 4 cores · 8 GB RAM |
Want to upgrade? Talk to an admin from contact, or buy credits and ask for promotion to super or ultra.