Prepare for GPU rental and your first run
What to include about your model, software environment, dates and test criteria.
01. Describe your workload
Include your model link, workload type, weight format and target context length. For inference, add expected concurrency; for fine-tuning, include dataset size and method. If you are still choosing a model, describe the outcome you need.
02. Prepare your environment
Save your dependencies, runtime version and launch command. Check ARM64 support for GX10. Prepare a small test dataset without sensitive information so compatibility checks and debugging are easy to repeat.
03. Define a successful test
Choose your metrics in advance: peak memory, processing time, output quality or experiment cost. Include preferred dates and duration in your request. Availability, configuration and access terms need confirmation before the session begins.
Need GPU rental for this workload?
Provide your model and preferred dates. We will confirm the configuration, rate, availability and connection method before the rental starts.