San Sebastian, Spain 16 March 2026 Multiverse Computing today announced it will soon host the newly announced Nemotron-3 family of models, including the newly announced Nemotron 3 Omni model, within its CompactifAI API. This gives organizations worldwide easier access to powerful multimodal generative AI through a scalable, cloud-based, developer-friendly platform.
By bringing the Nemotron-3 family of models to Multiverse Computing’s API, the company aims to make enterprise AI adoption more accessible while supporting organizations looking to evaluate and deploy high-performance language models in production environments.
Through the CompactifAI API, customers will gain immediate access to the NVIDIA’s Nemotron-3 Omni models once it is released and will also be able to benefit from future compressed versions, leveraging Multiverse Computing’s expertise in AI model compression. This approach enables organizations to maintain strong model performance while improving efficiency and reducing computational requirements. Multiverse Computing is also a member of the NVIDIA Inception program for startups, which supports companies advancing AI and accelerated computing.
According to Enrique Lizaso, Cofounder & Chief Executive Officer at Multiverse Computing, “By hosting NVIDIA’s Nemotron 3 family of models, including the upcoming next-generation multimodal Nemotron-3 Omni models on the CompactifAI API, we enable organizations to unlock a variety of industry use cases. Combined with scalable cloud access, this empowers companies to experiment, deploy, and scale generative AI faster than ever.”
Organizations interested in early access joining the waiting list for the upcoming launch can register their interest and receive updates through the CompactifAI API landing page: multiversecomputing.com/nemotron3
Interested developers and organizations who sign up during the GTC event will receive free credits for a limited time to start exploring and experimenting with these models as soon as they are available.
About Multiverse Computing
Multiverse Computing, headquartered in Donostia–San Sebastián (Spain) with offices in the United States, Canada and Europe, is a leader in compressed AI models. Its CompactifAI compression technology, delivers AI models compressed by up to 80% with minimal loss in accuracy, reducing computing requirements and enabling new AI applications. Learn more at www.multiversecomputing.com