Yeah, on-premise AI is another term for running the AI agent on your local computer or a server that you run and manage. This is the most private way to do AI because you don't go to the internet at all.

The trade-off is that your personal hardware is less powerful than the cloud, so you run smaller AI that sacrifices accuracy to speed.

WIth nostr:nprofile1qyjhwumn8ghj7en9v4j8xtnwdaehgu3wvfskuep0dakku62ltamx2mn5w4ex2ucpxpmhxue69uhkjarrdpuj6em0d3jx2mnjdajz6en4wf3k7mn5dphhq6rpva6hxtnnvdshyctz9e5k6tcqyp7u8zl8y8yfa87nstgj2405t2shal4rez0fzvxgrseq7k60gsrx6zeuh5t we are bringing as much privacy as we can into the cloud by using private encryption keys and secure enclaves. That means you can use the large computing power of the cloud to run the largest, most accurate open source models.

Local AI: Most Private, Slower, Lower Accurate, Offline support

Maple AI: Close to Local Privacy, Faster, Higher Accuracy, Online required