Memory for the model
Account for weights, context and cache with your chosen model format.
WORKLOADS / INFERENCE
GPU for inference
Choose a GPU for running your model. Consider weight size, context length, concurrent requests and budget. Benchmark your actual workload before scaling.
- Inference
BEFORE YOU LAUNCH
A configuration that fits your work
Measure latency
Test real requests. GPU specifications alone do not guarantee response times.
Budget for demand
Compare the hourly rate and runtime with the request volume you expect.
FIND YOUR CONFIGURATION
GPU rental & pricing
Hourly billing · USD
Checking configurations and prices…
GPU INSTANCE
/hour
- VRAM / GPU
- vCPU
- RAM
- Storage
- Region
Choose a configuration in the catalog. Check the current rate and terms in the dashboard before launch.
Full catalogFROM SELECTION TO LAUNCH
What will your run cost?
Choose your GPU, set up the environment and estimate your budget. All here, step by step.
Starting requires enough balance for 1 hour. Reviewing price and balance does not launch an instance: final confirmation is separate.
- 01 / Configure
- 02 / Settings
- 03 / Estimate
YOUR NEXT STEPS
From requirements to a first test
Define the requirements
Start with model size, GPU memory, storage and the software you plan to use.
Compare available offers
Check current prices and availability in the catalog, then review the configuration in your dashboard.
Configure and test
Prepare your environment and run a small test before committing to a longer workload.
FAQ
Questions before you start
Where can I check current prices?
Use the catalog and verify the final configuration and price in your dashboard before launching.
Is the software already installed?
This page does not guarantee a preinstalled environment. Check the selected image and install or configure your software when needed.