AI Inference

Ask the SolarCompute Network

Ask a question and watch your request run on an available SolarCompute machine.

Process

What happens next

01

Submit

The workload is submitted through the public launcher.

02

Validate

The request is checked against public workload rules.

03

Queue

The workload is queued for an available machine.

04

Run

The workload runs under limits and returns a status or result.

AI Inference

About this workload

AI inference workloads use an existing model or template to process a prompt, instruction, image, or other supported input.

These workloads are useful for testing prompt execution, machine allocation, runtime limits, queue handling, and result delivery across the SolarCompute network.

Supported public tests

  • Text inference
  • Summarisation
  • Question answering
  • Vision and image understanding
  • Prompt evaluation