NVIDIA Ising Enables Fully Automated Quantum Computer Calibration with Enhanced In-Context Learning
Quick Answer
NVIDIA's Ising Calibration 1.5 model automates quantum processor calibration, achieving 86.68% better performance than its predecessor on the QCalEval benchmark, while being 11.4% smaller.
Quick Take
It supports deployment on consumer GPUs and is fully open-source, allowing for customizable quantum calibration workflows.
Key Points
- Ising Calibration 1.5 analyzes diagnostic outputs without prior training examples.
- The model is 11.4% smaller and supports NVFP4 quantization for efficient deployment.
- It outperforms all open models and competes with state-of-the-art closed models.
- Performance evaluated using QCalEval benchmark for zero-shot and in-context learning.
- Open-source resources include model weights and deployment recipes for customization.
DeepSignal Analysis
What happened
NVIDIA has released Ising Calibration 1.5, an open-source model designed for automating quantum processor calibration. This model improves performance by 86.68% over its predecessor on the QCalEval benchmark and is 11.4% smaller, facilitating deployment on consumer GPUs.
Key evidence
- Ising Calibration 1.5 is a vision language model that interprets diagnostic outputs from quantum processors and suggests tuning adjustments.
- The model is evaluated using the QCalEval benchmark, which assesses its ability to analyze experimental results and recommend next steps.
- Ising Calibration 1.5 is available in a quantized version for deployment on consumer GPUs, making it accessible for local lab environments.
Why it matters
The advancements in Ising Calibration 1.5 could significantly streamline the calibration process for quantum processors, potentially leading to more efficient quantum computing operations. Its open-source nature allows researchers and developers to customize workflows, which may accelerate innovation in quantum technologies.
What to watch
📖 Reader Mode
~4 min readNVIDIA Ising Calibration is an open source vision language model (VLM) designed to interpret diagnostic outputs from quantum processors and determine how they should be tuned to continue operating.
This post introduces the latest model release, NVIDIA Ising Calibration 1.5, which advances AI-based QPU calibration by analyzing unfamiliar diagnostic results without prior training examples. Ising Calibration 1.5 also uses examples from related experiments when available and is 11.4% smaller at BF16 precision. This eases the deployment of agentic calibration workflows directly in local lab environments.
For the first time, the model is also available in an NVFP4-quantized version, enabling deployment on a single GPU or an NVIDIA DGX Spark—comparable with leading closed models such as Fable 5 and GPT 5.6 Sol.
How is the Ising Calibration 1.5 model trained?
The Ising Calibration 1.5 model is trained on data generated from partner contributions across multiple qubit modalities, including superconducting qubits, quantum dots, ions, neutral atoms, electrons on Helium, and others specializing in calibration and control.
How is Ising Calibration 1.5 performance evaluated?
Performance of Ising Calibration 1.5 is evaluated using the QCalEval benchmark, which measures a model’s ability to interpret experimental results, classify outcomes, evaluate significance, assess fit quality and key features, and recommend next steps. For additional details on the benchmark, model architecture, and evaluation results, see QCalEval: Benchmarking Vision-Language Models for Quantum Calibration Plot Understanding.
The evaluation covers both zero-shot and in-context learning (ICL). Zero-shot reasoning analyzes results independently, while ICL evaluates results in the context of related samples. Both are important for building agents that can automate QPU bring-up and retune operations.
On the QCalEval benchmark, Ising Calibration 1.5 shows strong performance when analyzing diagnostic results without prior examples. It is now also 86.68% better than its predecessor when using examples from related experiments. It outperforms comparable open models and remains competitive with leading closed models.
Ising Calibration 1.5 advances AI and quantum computing calibration by outperforming all open models out of the box and on the QCalEval benchmark. It is competitive with state-of-the-art closed or 1T+ parameter models.
The 31-billion-parameter VLM is suited for data center GPUs such as NVIDIA Grace Blackwell and NVIDIA Vera Rubin. It also ships with a quantized version to NVFP4 where users can run this model on a consumer gaming card or NVIDIA DGX Spark with only a small cost in accuracy.
The tokens per second (TPS) performance on DGX Spark has also been optimized, enabling good throughput locally for a fraction of the cost.
To learn more about deploying Ising Calibration 1.5 with an agent, check out the NVIDIA/Quantum-Calibration-Agent-Blueprint GitHub repo.
Get started with NVIDIA Ising open resources
The NVIDIA Ising model family is fully open. Weights, data, benchmarks, and recipes are provided so you can modify, deploy, and fine-tune your own models and variants for your specific QPUs.
Model weights
Full-parameter checkpoints for Ising Calibration 1.5 are available on Hugging Face:
Ising Calibration 1.5 is also available as an NVIDIA NIM and hosted through NVIDIA Build. The OpenMDW License from Linux Foundation offers QPU builders and operators with the flexibility to maintain data control and deploy anywhere.
Deployment recipes
A ready-to-use agent harness blueprint is available with support for this model and others, including large model cloud APIs.
- Quantum calibration agent blueprint is a script for deploying an agentic workflow using Ising Calibration 1.5 with the NVIDIA Nemo Agent Toolkit to quickly set up quantum calibration experiment automation.
Open datasets and QCalEval benchmark
Ising Calibration 1.5 is built on real QPU data provided by partners and collaborators. A semantic quantum calibration benchmark has also been released to evaluate model effectiveness for this task.
- Check out the QCalEval dataset
- Run the QCalEval script
About the Authors
— Originally published at developer.nvidia.com
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from NVIDIA Developer Blog
See more →
Synthetic Data Generation for Financial AI Research with NVIDIA NeMo
NVIDIA's NeMo pipeline generates 502,536 unique financial news headlines in 82 iterations, addressing data imbalance in financial NLP. The iterative approach uses semantic deduplication and category-weighted sampling to enhance diversity and relevance in generated content.
