NVIDIA Sees Open AI Models Driving a Multi-Model Enterprise Future

NVIDIA expects a multi-model enterprise AI future, where open-source and closed frontier models work together rather than replace one another.

Smaller customized models may handle targeted tasks, while frontier models remain useful for unfamiliar or complex workloads.

Open models can help enterprises fine-tune AI on proprietary data, improving performance for specific use cases while offering greater control, lower costs and lower latency—especially for edge and sensitive workloads.

NVIDIA sees model routers and agent “harnesses” as increasingly important for directing workloads across local hardware, private infrastructure, cloud services and frontier models while managing security, data privacy and performance.

Director of Developer Tech Nader Khalil said open-source AI models are helping broaden access to model customization, accelerate development and support workloads that require local deployment, while closed frontier models will remain part of a “multi-model world.”

Speaking at The Six Five Summit 2026, Khalil described open source as a longstanding catalyst for innovation across software categories, rather than an AI-specific concept. He said the degree to which an AI model is open can vary widely, including whether developers receive access to a model’s weights, architecture or training data.

Khalil pointed to the evolution of models including RedPajama, Stable LM and Meta’s Llama releases as examples of the industry determining where value resides in AI development. In his view, model weights can be relatively transient, while data is a more durable asset.

“Value was accruing at the data,” Khalil said, arguing that enterprises hold substantial proprietary information that may not be sufficient or economical to use for training a foundation model from scratch. Instead, organizations can post-train or fine-tune an existing model on their own data for specific tasks.

He said a smaller customized model can be faster, less resource-intensive and less expensive for work that falls within a defined domain. However, he said frontier models can remain appropriate for more open-ended or unfamiliar tasks.

Khalil offered his own use of a DGX Station as an example, saying he runs GLM-5.2 locally at roughly 100 tokens per second and uses it alongside Claude and ChatGPT. He said local or open models do not necessarily replace closed models; rather, organizations may use each for different parts of a workflow.

Customized smaller models can handle targeted, in-domain tasks. Frontier models can be used for tasks outside a defined domain. Locally deployed models can support workloads where latency or control over data is critical.

Khalil said NVIDIA releases models, datasets and tooling, including its Nemotron family, because the company sees a bottleneck in enabling enterprises to apply their data to capable, fully open models. He said NVIDIA’s approach includes releasing more than model weights, including datasets that users can adapt or use to train their own models.

He described model selection as a tradeoff between intelligence and speed. Some applications can tolerate slower processing in exchange for greater capability, while others require low latency and may be better served by a smaller, faster model. The appropriate choice depends on the use case, he said. For workloads at the edge, Khalil said access to model weights is nece...