All terms
Glossary
Inference
Inference is using a trained model to produce an answer, as opposed to training it. Every prompt you send to a model is an inference request. Where inference runs, and on what hardware, decides response time, data exposure and the way you pay for it.
Definitions are general and kept short on purpose. Check the vendor documentation or the regulation for the details of your case.
Next step
Want to talk it through?
If a term keeps coming up in your project, we can explain what it means for your case.
Currently accepting projects for Q4 2026