# NVIDIA Links Quantum Models to Processor and Cloud Workflows

By Simon Yoon

Canonical URL: https://www.tokenpost.com/news/technology/37220
Published: 2026-10-22T00:19:30.000Z
Updated: 2026-10-22T00:19:30.000Z
Section: Technology

> The Oct. 21 GTC Berlin update also spans autonomous driving, protein design and enterprise data, with products on different release schedules.

NVIDIA tied its Ising quantum models to processor calibration and cloud computing at GTC Berlin on Oct. 21, while expanding its AI lineup across vehicles, protein design and enterprise data.

Ising integrations offer one example of the lineup’s specialized computing uses. IQM Quantum Computers is using Ising for quantum processor calibration. SEEQC is combining the models with NVIDIA NVQLink, while Amazon Web Services (AWS) is integrating Ising into AWS Parallel Computing Service. The models are designed for calibration and error correction.

For physical AI, Cosmos Data Factory brings together tools to curate, label, generate and assess training data, including data for driving scenarios and product defects. Alpamayo 2 Edge is a compact model for real-time vehicle scene understanding and motion planning, optimized for NVIDIA DRIVE AGX Thor. It is expected later in October.

NVIDIA’s BioNeMo RFD4-Proteina, co-developed with the University of Washington’s Institute for Protein Design, generates protein sequences and three-dimensional structures. “RFD4-Proteina will open the door to new solutions in medicine, sustainability, materials and biotechnology,” said David Baker, director of the institute and a professor of biochemistry at the University of Washington School of Medicine.

The enterprise lineup includes Nemotron Retriever models for embedding, reranking and document parsing. Kumo models are designed for prediction, forecasting and anomaly detection using relational and tabular data. Coinbase is among Kumo’s early users and is [working with the model family](<https://www.tokenpost.com/news/technology/37039>).

The portfolio combines products at different stages. Nemotron 3.5 Super VL, a 120 billion-parameter model, was announced Oct. 15. Nemotron Bolt E4B 4GB is scheduled for Oct. 23. Nemotron Retriever, Kumo, Cosmos, BioNeMo RFD4-Proteina and Ising are available through varying combinations of Hugging Face, GitHub and build.nvidia.com.



## Links in this article

- [working with the model family](https://www.tokenpost.com/news/technology/37039)
