How to Rent Your First GPU
Step-by-step guide to renting a GPU on GPUniq — from account creation to running your first workload.
Quick Start
Create Account
Go to gpuniq.com and sign up with your email or Google account. Verify your email with the 6-digit code.
Add Funds
Navigate to Balance in your dashboard. Choose your deposit method: USDT (TRC20), Stripe, or YooKassa.
Get API Key
Go to LLM API Keys and create a new key starting with gpuniq_.
Install SDK
pip install GPUniq
Deploy
from gpuniq import GPUniq
client = GPUniq(api_key="gpuniq_your_key")
deploy = client.gpu_cloud.deploy(
gpu_name="RTX_4090",
docker_image="pytorch/pytorch:latest",
disk_gb=100,
)
Connect
Find SSH credentials in the dashboard:
ssh root@{host} -p {port}
After Connecting
Verify GPU Access
nvidia-smi
nvcc --version
python -c "import torch; print(torch.cuda.is_available())"
Install Your Tools
pip install torch torchvision transformers
Transfer Your Data
# Upload data to a volume
client.volumes.upload(volume_id=1, file_path="./data.tar.gz")
# Attach volume when deploying
deploy = client.gpu_cloud.deploy(
gpu_name="RTX_4090",
volume_id=1,
)
scp -P {port} my_data.zip root@{host}:/root/
rsync -avz -e "ssh -p {port}" ./project root@{host}:/root/
Managing Your Rental
# Stop billing (pause instance)
client.instances.stop(task_id=456)
# Resume
client.instances.start(task_id=456)
# Delete permanently
client.instances.delete(task_id=456)
Use Volumes to persist data between instances. Instance storage is lost on delete.
Next Steps
- Explore the GPU Marketplace for more hardware options
- Set up SSH Keys for passwordless access
- Read the full SDK Reference
Was this page helpful?
Last updated Feb 22, 2026
Built with Documentation.AI