Local Models: Trust, Control, Optimization — Carter Abdallah, NVIDIA

The argument in this local-model discussion is that dependable access can matter as much as benchmark performance. Running an open model gives an organization control over its own deployment, while access through another company's service remains dependent on that provider.

The panel also argues for adapting models to a narrow workload instead of judging them only by broad rankings. These are reasons an organization might value ownership and customization. The customer examples are participants' accounts, so they support a deployment argument without establishing that local models consistently outperform hosted alternatives.

“an open model is a directory of files you can inspect, running on code you can read, while the same claim about a closed API is unverifiable by construction”

“Closed terms of service bar you from training on outputs, so owning the model also means owning the traces, which is what makes a data flywheel possible at all”

“a model tuned to be good across every harness is optimal for nobody's, and open weights let you fit it to the one or two things you actually do”

“enterprises move to Chinese open models, not because they scored better but because availability could be counted on”