RTX 4090 · Now available

GPU compute,
on demand.

Launch a JupyterLab backed by an NVIDIA RTX 4090 in seconds. Built-in auth, Stripe billing, and a full REST API. From .39/hr — no credit card to start.

Start for free → View API docs

RTX 4090 · 24 GB VRAM CUDA 12.1 · PyTorch · Transformers Stripe billing · Cancel any time

Quickstart — 3 commands
# 1. Register
curl -X POST https://gpu.nexaquantum.net/auth/register \
  -H "Content-Type: application/json" \
  -d '{"email":"you@example.com","password":"StrongPass1!",
     "full_name":"Your Name","username":"yourhandle"}'

# 2. Login → receive JWT
curl -X POST https://gpu.nexaquantum.net/auth/login \
  -d "email=you@example.com&password=StrongPass1!"

# 3. Launch a GPU lab
curl -X POST https://gpu.nexaquantum.net/labs \
  -H "Authorization: Bearer <your-token>" \
  -d '{"name":"mylab","gpus":1,"port":8888}'
Full reference → gpu.nexaquantum.net/docs

What you get

Everything you need to run GPU workloads

No infrastructure management. GPU-backed compute accessible via API, ready in seconds.

⚡

Instant JupyterLab

GPU-backed notebook ready in under 30 seconds. Pre-loaded with PyTorch, CUDA 12.1, Hugging Face Transformers, and more.

🔒

Built-in Auth

JWT + API key authentication included. Every lab runs in an isolated Docker container — only you have access.

💳

Transparent Billing

Prepaid monthly plans via Stripe. Automatic usage metering, email invoices. Cancel any time — no contracts, no surprises.

🔧

Full REST API

Create, list, and delete labs programmatically. Complete OpenAPI docs at /docs. Integrate into your scripts or CI/CD pipelines.

💾

Persistent Storage

Your notebooks and data survive container restarts. Up to 50 GB on paid plans. Download your files any time via JupyterLab.

📊

Usage Tracking

Monitor compute hours, storage, and billing in real time via the API. No hidden charges — you see exactly what you're spending.

Get started

Up and running in under a minute

Create a free account below or via POST /auth/register. No credit card required for the free tier.

Get your token

Launch a lab

Call POST /labs with your token. A JupyterLab container with GPU access starts in seconds.

Scale when ready

Upgrade to a paid plan for more GPU hours, longer sessions, and priority scheduling.

Pricing

Simple, transparent pricing

Start free. Pay only for what you use. No hidden fees or long-term commitments.

Free

Try it out with no commitment.

/ month

No credit card required

2 GPU hours
RTX 4090 · 24 GB VRAM
10 GB storage
Max 2-hour sessions
Full API access

Get started →

Start free — no credit card needed

2 free GPU hours on an RTX 4090
Full REST API access from day one
Persistent storage — your data is preserved
Simple per-hour billing, cancel any time

Create your account

Free tier · No credit card required.

Already have an account? Sign in via API →

FAQ

Frequently asked questions

Do I need a credit card to start?

No. The free tier gives you 2 hours on an RTX 4090 with no payment method required. Add a card only when you're ready to upgrade.

How does billing work?

Paid plans are prepaid monthly bundles billed via Stripe. Starter is 40 GPU hours at .60/mo (.39/hr). Pro is 100 hours at /mo (.29/hr). Unused hours do not roll over.

Can I use the platform programmatically?

Yes — the entire platform is API-first. Create labs, check GPU status, and manage billing via REST. Complete OpenAPI documentation is available at /docs.

Is my data private?

Yes. Each lab runs in an isolated Docker container with its own file system. Other users cannot access your session or your data.

What happens when a session ends?

Your container stops, but your persistent storage is preserved. Restart your lab any time to continue where you left off.

Can I export my work?

Yes. Download notebooks, datasets, and checkpoints any time via the JupyterLab file browser. You own your data.