Glossary · Model mechanics
Inference
Also said as
running a modelmodel hostinghosted inference
Inference is the act of running a trained AI model — every time you send a prompt, something somewhere is doing the inference.
Why it matters to your business
Somebody pays for that compute: cloud tools bake it into the subscription, and self-hosted tools move it onto your own hardware — which is exactly the trade in this directory.
The catch · what vendors don't say
Even 'free forever' local tools sell hosted inference as an upgrade — the machine has to run somewhere, and running it is the cost.
