NVIDIA Ising launched as the world’s first family of open source quantum AI models, targeting the two biggest engineering bottlenecks in quantum computing: processor calibration and error-corrected decoding.
Summary
- NVIDIA Ising delivers up to 2.5x faster and 3x more accurate quantum error correction decoding than current open source benchmarks, and calibration workflows are reduced from days to hours.
- The model family includes Ising Calibration, a 35 billion parameter vision language model, and Ising Decoding, a 3D convolutional neural network framework, both available on GitHub and Hugging Face.
- Early adopters include Fermi National Accelerator Laboratory, Harvard, IQM Quantum Computers, Lawrence Berkeley National Laboratory and the UK National Physical Laboratory.
NVIDIA Ising launched April 15, 2026, as the world’s first family of open source AI models designed specifically for quantum computing, providing researchers and companies with tools to address processor calibration and error correction, the two engineering barriers that stand between today’s fragile qubits and useful large-scale quantum systems.
The models achieve up to 2.5x faster and 3x more accurate quantum error correction decoding compared to pyMatching, the current open source benchmark.
The family has two domains. Ising Calibration is a 35 billion parameter vision language model that automates quantum processor tuning, compressing calibration workflows that previously required days of manual configuration to hours of automated execution. Ising Decoding is a 3D convolutional neural network framework for real-time quantum error correction, available in two variants optimized for speed or accuracy depending on the application.
Both models are distributed through GitHub, Hugging Face and NVIDIA’s build.nvidia.com platform, integrated with CUDA-Q and NVQLink. NVIDIA is also releasing a quantum workflow cookbook, training data sets, and hardware-specific tuning tools so researchers can adapt the models to their own quantum processor architectures without exposing proprietary data.
Jensen Huang, founder and CEO of NVIDIA, framed the launch in terms of infrastructure. “AI is essential to making quantum computing practical. With Ising, AI becomes the control plane, the operating system of quantum machines, transforming fragile qubits into scalable and reliable quantum GPU systems,” he said.
Who is already using it?
Adoption at launch spans a variety of institutions, including Academia Sinica, Fermi National Accelerator Laboratory, Harvard John A. Paulson School of Engineering and Applied Sciences, IQM Quantum Computers, Lawrence Berkeley National Laboratory’s Advanced Quantum Testbed, Sandia National Laboratories, the University of California, San Diego, the United Kingdom’s National Physical Laboratory, and Yonsei University.
The breadth of early adopters reflects a deliberate open model strategy. By publicly publishing pre-trained weights, training frameworks, and benchmarks, NVIDIA positions Ising as a base layer that other developers can build on without starting from scratch.
Implications of the cryptocurrency and artificial intelligence market
The launch of Ising reinforces NVIDIA’s position as the dominant infrastructure provider in both classical AI and the emerging hybrid quantum-classical computing stack. For the crypto sector, quantum computing has long represented a future threat to existing blockchain encryption standards, particularly RSA and the elliptic curve cryptography used to secure Bitcoin wallets.
Progress in quantum error correction, which Ising specifically addresses, is the technical precondition for cryptographically relevant quantum computers to exist. The timeline is still far away, but every improvement in error-correction decoding accuracy shortens it.
NVIDIA news has historically triggered moves in AI tokens throughout the cryptocurrency market, as the chip company’s hardware underpins the AI infrastructure that powers many blockchain AI projects. The launch of Ising adds a new quantum AI vertical to that relationship.
