Getting Started

Complete guide to get your first GPU instance running on Spheron. This guide takes you from account creation to a deployed instance in about 10 minutes.

Already have an account? Jump to Quick Start for fast deployment.

What You'll Accomplish

By the end of this guide, you will:

• ✅ Create and configure your Spheron AI account
• ✅ Add billing and credits
• ✅ Generate and upload SSH keys
• ✅ Deploy your first GPU instance
• ✅ Connect and verify your instance

Step 1: Create Your Account

Sign Up

1. Visit app.spheron.ai
2. Click Sign Up
3. Choose sign-up method:
   • GitHub account (recommended for developers)
   • Email and password
4. Complete email verification if using email signup

Navigate the Dashboard

Once logged in, you'll see:

• Deploy - Create new GPU instances
• Instances - Manage active deployments
• Credits - Add funds and view balance
• Settings - Account and SSH key management

Step 2: Add Billing

Add Credits

1. Click Credits in the top-right corner
2. Choose payment method:
   • Credit/Debit Card
   • Cryptocurrency
3. Add initial credits (minimum varies by payment method)

Pricing: Pay-per-second billing with no hidden fees. See Billing for detailed pricing.

Tip: Start with $10-20 for testing. RTX 4090 costs ~$0.52/hour.

Step 3: Set Up SSH Access

SSH keys are required to access your GPU instances securely.

Generate SSH Key

# Generate new SSH key
ssh-keygen -t ed25519 -C "your_email@example.com"
 
# Press Enter to save to default location
# Set a passphrase (optional but recommended)
 
# Display your public key
cat ~/.ssh/id_ed25519.pub

# Generate new SSH key
ssh-keygen -t ed25519 -C "your_email@example.com"
 
# Display your public key
type $env:USERPROFILE\.ssh\id_ed25519.pub

Upload SSH Key to Spheron

1. Go to User Settings
2. Navigate to SSH Keys section
3. Click Add SSH Key
4. Paste your public key (the one ending in .pub)
5. Give it a memorable name (e.g., "My Laptop")
6. Click Save

Step 4: Deploy Your First Instance

Navigate to Deploy

1. Click Deploy in the left sidebar
2. Browse the GPU catalog

Select Your GPU

• GPU: RTX 4090 (24GB VRAM)
• Cost: ~$0.52/hour
• Good for: Learning, testing, small models

Click on the RTX 4090 to proceed.

Configure Instance

1. Region: Choose the one closest to you

   • Lower latency
   • Better performance

2. Operating System: Select Ubuntu 22.04 LTS

   • Most stable
   • Best GPU driver support

3. Storage: Keep default (usually sufficient)

4. SSH Key: Select the key you uploaded earlier

5. Startup Script (Optional): Leave blank for now

   • See Startup Scripts for advanced automation

Review and Deploy

1. Check the Order Summary on the right:

   • Hourly cost
   • Your current balance
   • Instance configuration

2. Click Deploy Instance

3. Wait 30-60 seconds for provisioning

Step 5: Connect to Your Instance

Get Connection Details

Once deployed, your instance dashboard will show:

• IP address
• Username (usually root, ubuntu, or provider-specific)
• SSH connection command

Connect via SSH

ssh root@123.456.789.012

ssh -i ~/.ssh/id_ed25519 root@123.456.789.012

• You'll see a fingerprint verification message
• Type yes to continue
• Enter your SSH key passphrase if you set one

Step 6: Verify Your Setup

Once connected, verify everything is working:

Check GPU

nvidia-smi

• GPU model (RTX 4090)
• Memory (24GB)
• Driver version
• GPU utilization (should be low when idle)

Check CUDA

nvcc --version

Should show CUDA compiler version.

Check System

# Check CPU and memory
htop
 
# Check disk space
df -h
 
# Check OS version
cat /etc/os-release

Test GPU Access

# Simple GPU test
python3 -c "import torch; print(f'CUDA available: {torch.cuda.is_available()}')"

If PyTorch isn't installed, that's fine - you'll install it when needed.

What's Next?

Congratulations! You've successfully deployed your first GPU instance. Here's what to explore next:

Quick Deployments

• Quick Start - Fast deployment with templates and examples
• Deploy LLMs - Run specific AI models
• Reserved GPUs - Lock in GPUs for better rates

Connect Different Ways

• SSH Connection Guide - Advanced SSH setup
• Jupyter Notebook - Browser-based development
• VS Code Remote - Code directly on your instance

Platform Features

• API Reference - Automate deployments
• Security Best Practices - Keep your account secure
• Billing - Understand costs and payments

Troubleshooting

Cannot connect via SSH

# Verify your SSH key is loaded
ssh-add -l

ssh-add ~/.ssh/id_ed25519

• Verify you uploaded the correct public key
• Check firewall rules allow SSH (port 22)
• Try with explicit key: ssh -i ~/.ssh/id_ed25519 user@ip

GPU not detected

• Run nvidia-smi - should show GPU details
• If error, wait 30 seconds and try again (drivers may be loading)
• Reboot instance: sudo reboot

Instance deployment failed

• Check your account balance has sufficient credits
• Try a different region (some may be at capacity)
• Contact support if issue persists

Managing Your Instance

Terminate Instance

When done with your instance, terminate it to stop charges:

1. Go to instance dashboard
2. Click Terminate or Delete
3. Instance stops and all data is deleted permanently

Important: Always backup important data before terminating. Once terminated, your instance and data cannot be recovered.

Need Help?

• Documentation: Browse all guides
• Community: Join Discord
• Support: Visit General Info for official channels
• Security: Review Security Best Practices

Ready for faster deployments? Check out the Quick Start guide for streamlined workflows.