Nvidia and French startup Mistral AI have collaboratively announced the release of a new language model dubbed Mistral-NeMo, which is set to bring powerful AI capabilities directly to business desktops.

The model possesses 12 billion parameters and an expansive 128,000 token context window, This makes it a daunting tool for businesses seeking to implement AI solutions without the need for extensive cloud resources.

In an interview with VentureBeat, Bryan Catanzaro, VP of Applied Deep Learning Research at Nvidia, highlighted the model’s accessibility and efficiency. He said, “We’re launching a model that we jointly trained with Mistral. It’s a 12 billion parameter model, and we’re launching it under Apache 2.0,” he said. “We’re really excited about the accuracy of this model across a lot of tasks.”

Elaborating on the model’s advantages, Catanzaro said, “The smaller models are just dramatically more accessible.” He added, “They’re easier to run, the business model can be different because people can run them on their own systems at home. In fact, this model can run on RTX GPUs that many people have already.”

With the 128,000 token window feature, Mistral-NeMo can process and understand massive chunks of text compared to its competitors. Notably, the extended window could prove significantly valuable for businesses dealing with lengthy documents, complex analyses, or intricate coding tasks.

Clarifying the model’s intended use case, Catanzaro says that he would like to think more about laptops and desktop PCs than smartphones. The statement hints that Mistral-NeMo is not yet ready for mobile deployment.

The model is currently available as an NVIDIA NIM inference microservice, with a downloadable version expected in the near future.