Topic

On-premises inference

Serving trained AI models on hardware an organisation owns and operates itself instead of renting cloud capacity.

Current clusters

product3 publishers

Apple weighs wiring an M8 Ultra AI server together with Nvidia's NVLink Fusion

The Information reports two configurations, one with two chips and one with four, aimed at developers, businesses and governments, with a possible 2029 launch. Apple already builds servers for Private Cloud Compute and has turned partners away from them.

Publishers:9to5mac.commacrumors.comthenextweb.com

Reality

Evidence32
Adoption18
Hype gap+35
Incentives50
Confidence55