Cassava Deploys NVIDIA Nemotron Through CAIMEx

Cassava Technologies is deploying NVIDIA Nemotron models through the Cassava AI Multi Model Exchange, or CAIMEx, as a locally hosted managed inference service, the company said in a press release. Customers will be able to access the open models through APIs without operating their own model serving infrastructure.

Nemotron provides reasoning and tool use capabilities for building AI agents that interact with business systems. Developers can also customize the models using their open weights, training data and recipes.

Cassava is also adding NVIDIA OpenShell to support governed agent execution and separate security enforcement. Its CAIMEx Token Optimiser identifies previously processed information and reuses it when appropriate, reducing repeated computing across Cassava hosted and third party models.