What is VoltageGPU API?

VoltageGPU provides two APIs: a GPU Pods API for deploying and managing GPU instances, and an AI Inference API for running 140+ AI models including DeepSeek-R1, Llama 3, and FLUX.

Is VoltageGPU API OpenAI compatible?

Yes, the VoltageGPU AI Inference API is fully OpenAI-compatible. You can use existing OpenAI SDKs by simply changing the base URL to https://api.voltagegpu.com/v1

What AI models are available?

VoltageGPU offers 140+ AI models including DeepSeek-R1, DeepSeek-V3, Llama 3.3 70B, Qwen 2.5, Mixtral, FLUX.1, Stable Diffusion, and many more for chat, embeddings, image and video generation.

Documentation

Volumes

Volumes in VoltageGPU provide persistent storage for your GPU rental containers, ensuring your important data remains safe even if your rental pod is unexpectedly terminated.

Overview

Volumes are backed by secure cloud storage and can be attached to any of your pods, giving you reliable data persistence across different rental sessions.

When you attach a volume to your pod, it becomes available at the /mnt directory inside your container, allowing you to easily store and retrieve your data, models, datasets, and other important files.

How It Works

VoltageGPU volumes are implemented using cloud storage technology that seamlessly integrates with your rental containers. When you create a volume, it's allocated in our secure cloud storage infrastructure and can be mounted to any of your active pods.

Key Features

Persistent Storage

Your data remains available even when pods are terminated.

Cross-Pod Access

Attach the same volume to different pods as needed.

Automatic Mounting

Volumes are automatically mounted at /mnt when attached to a pod.

Scalable

Volumes automatically scale with your storage needs.

Getting Started

Step 1: Create a Volume

Navigate to the VoltageGPU dashboard
Click "Access" at the bottom of sidebar
Go to the "Volumes" section
Click "Create Volume"
Specify your volume name and description
Click "Create" to provision your volume

Your volume will be created and ready to attach to any of your pods.

Step 2: Attach Volume to Pod

When deploying a new pod:

In the pod deployment configuration
Select your created volume from the "Volumes" dropdown
Complete your pod deployment as usual

Once your pod is successfully deployed, the volume will be automatically mounted at /mnt inside your container.

Accessing Your Volume

After your pod starts with an attached volume:

# Navigate to your volume
cd /mnt

# List contents
ls -la

# Your persistent data is available here

Billing

Volume billing is based on actual storage usage and is charged hourly:

Storage Cost: Billed based on the amount of data stored
Always Active: Volumes incur charges even when not attached to a pod
Pay-as-you-use: Costs scale with your actual storage usage
Transparent Pricing: Real-time billing updates in your dashboard

⚠️ Important Billing Note

Volumes continue to incur storage charges even when not attached to any pod. This ensures your data remains persistent and available whenever you need it.

Best Practices

Data Organization

# Recommended directory structure in /mnt
/mnt/
├── datasets/          # Input data and datasets
├── models/           # Trained models and weights  
├── checkpoints/      # Training checkpoints
├── outputs/          # Results and generated files
└── backups/          # Important file backups

Performance Tips

📝 Note

Volume operations may be slower than local storage as data is synced with external cloud storage.

Use volumes primarily for persistent data storage rather than high-frequency read/write operations
Store frequently accessed data locally during processing, then save results to volumes
Use volumes for checkpoint saving during long training runs
Keep important results on volumes to prevent data loss
Consider copying critical files from volumes to local storage for intensive processing tasks

Security

Volumes are private to your account and not accessible by other users
Regular backups are handled automatically by the platform
Data is encrypted at rest

Frequently Asked Questions

What happens to my volume if my pod crashes?

Your volume and all data remain completely intact. You can attach the same volume to a new pod and access all your files immediately.

Can I attach one volume to multiple pods simultaneously?

Currently, each volume can only be attached to one pod at a time. However, you can detach and reattach volumes to different pods as needed.

Is there a size limit for volumes?

Volumes automatically scale with your storage needs. You can store as much data as required without pre-defining size limits.

Am I charged for empty volumes?

You're only charged for the actual data stored on your volumes, not the allocated space.

Why am I being charged when my volume isn't attached to any pod?

Volumes maintain your data persistently, which requires ongoing storage infrastructure. This ensures your data is always available when you need it.

How can I reduce my volume costs?

Remove unnecessary files, compress large datasets, and delete unused volumes to optimize your storage costs.

Can I access my volume from outside the pod?

Volumes are only accessible from within your rental pods at the /mnt directory. This ensures security and proper access control.

My volume isn't showing up at /mnt. What should I do?

Ensure the volume was properly selected during pod deployment. If the issue persists, restart your pod or contact support.

Ready to Create Your Volume?

Persistent storage for your datasets, models, and checkpoints.

Create Volume Learn About Backups