Skip to content

Ubuntu Environments

Available Ubuntu configurations for GPU instances, optimized for AI/ML workloads.

Available Versions

VersionSupport UntilBest For
Ubuntu 20.04 LTS2025Legacy applications, older dependencies
Ubuntu 22.04 LTS2027Production workloads (most stable)
Ubuntu 24.04 LTS2029Latest features, experimental projects

Configuration Options

Base Images

  • Base / LTS Base - Minimal Ubuntu installation
  • + NVIDIA 550/570 - NVIDIA drivers pre-installed
  • + CUDA X.X - CUDA toolkit included
  • + Docker - Docker pre-installed for containerized workflows
  • Shade OS - Optimized lightweight version for maximum GPU performance

Pre-configured ML Environments (Ubuntu 24.04)

EnvironmentIncludesBest For
ML EverythingPyTorch, TensorFlow, JAXMulti-framework experimentation
ML PyTorchPyTorch optimizedLLM training, computer vision
ML TensorFlowTensorFlow optimizedProduction ML, enterprise

CUDA Versions Available

CUDA VersionFeaturesCompatibility
12.0Baseline, maximum compatibilityOlder frameworks
12.4Bug fixes, stableGeneral AI development
12.6Newer GPU optimizationsRTX 5090, H100
12.8 OpenOpen-source driversCommunity projects
13.0 OpenLatest featuresCutting-edge research

Selection Guide

Use CaseRecommended EnvironmentWhy
BeginnersUbuntu 24.04 ML EverythingAll frameworks pre-installed
LLM TrainingUbuntu 24.04 ML PyTorchPyTorch optimized
TensorFlowUbuntu 24.04 ML TensorFlowTensorFlow optimized
ProductionUbuntu 22.04 + CUDA 12.8 + DockerStable, containerized
ResearchUbuntu 24.04 + CUDA 13.0 OpenLatest features
Legacy AppsUbuntu 20.04 LTSOlder dependency support
Max PerformanceUbuntu 22.04 (Shade OS)Optimized, minimal overhead

Docker vs Non-Docker

Without Docker:
  • Direct GPU access
  • Simpler setup
  • Single-purpose instances
  • Good for: Learning, simple projects
With Docker:
  • Containerized workflows
  • Dependency isolation
  • Multi-project instances
  • Good for: Production, complex setups

Deploying

  1. Go to app.spheron.ai → Deploy
  2. Select GPU
  3. Choose Ubuntu environment from OS dropdown
  4. Deploy (ready in 30-60 seconds)

Verify Installation

After deployment, connect and verify:

# Connect
ssh root@your-instance-ip
 
# Check OS version
cat /etc/os-release
 
# Check CUDA (if applicable)
nvcc --version
 
# Check GPU
nvidia-smi
 
# Check Docker (if applicable)
docker --version

Frequently Asked Questions

What does LTS mean?
Long Term Support - 5 years of security updates and bug fixes.

Can I change environments after deployment?
No. Deploy new instance with desired environment.

Do I need Docker?
Not for simple projects. Use Docker for complex dependencies or multi-project instances.

Which CUDA version should I use?
CUDA 12.8 for best balance. Check framework compatibility first.

Can I install multiple CUDA versions?
Not recommended. Select correct version during deployment.

Ubuntu 22.04 or 24.04?
22.04 for production stability. 24.04 for latest features.

What is Shade OS?
Optimized Ubuntu variant with minimal overhead for maximum GPU performance.

Additional Resources

Powered by Spheron
© 2025 Spheron Network. All rights reserved.