NVIDIA + Infleqtion: 100 Logical Qubits And The 2028 Quantum Tipping Point

NVIDIA has partnered with Infleqtion to bridge the gap between traditional GPU-powered supercomputers and the burgeoning field of quantum computing.

Watch on YouTube For The Full experience - https://www.youtube.com/watch?v=AJjO6ZKcd5w&t=121s

At the heart of this collaboration is NVQ-Link, a low-latency, high-bandwidth universal interface designed to tightly couple quantum processing units (QPUs) with existing AI supercomputers. By achieving a staggering four-microsecond round-trip latency, this technology allows for real-time co-processing, enabling complex application workflows to be "sliced and diced" so that the most demanding quantum-native problems run on neutral atom qubits while parallel tasks remain on high-performance GPUs.

This hybrid architecture effectively transforms the quantum computer from an isolated lab instrument into a powerful, integrated co-processor capable of tackling previously unsolvable challenges in material science, drug discovery, and physical AI.

Mark and Jeremy sit down with Pranav Gokhale, CTO of Infleqtion, and Sam Stanwyck, Group Product Manager for Quantum Computing at Nvidia.

The road to commercial quantum utility is also being fast-tracked through the integration of AI-driven error correction and the development of logical qubits. In late 2024, the partnership achieved a historic milestone by demonstrating a material science application using the world's first logical qubits on a hybrid system.

Unlike the "noisy" and error-prone qubits of the past, logical qubits provide the pristine data necessary for reliable computation. With a roadmap aiming for 100 logical qubits by 2028, researchers anticipate a "tipping point" where quantum-enhanced systems will revolutionize industries by simulating electron-level interactions for next-generation batteries and aerospace materials.

As NVIDIA’s CUDA-Q software stack matures, the vision of a scientist simply prompting an AI orchestrator to solve deep physics problems is moving from science fiction to an imminent industrial reality

Chapters

(00:00) Why quantum computing matters right now

(01:20) Why Nvidia is betting big on quantum

(02:52) NVQ-Link: the bridge between quantum and classical computing

(09:29) Who decides what runs on the quantum computer vs the GPU?

(12:33) AI helping quantum, quantum helping AI

(16:56) Building a space elevator battery: a real quantum workflow

(20:09) The quantum algorithm zoo

(22:04) From noisy qubits to logical qubits

(24:00) How much energy does a quantum computer actually use?

(27:05) The no-cloning theorem: why you can't copy-paste quantum data

(27:20) The biggest unanswered question in quantum computing

(30:47) A $20M NASA program and a telescope for underground

(33:32) What do we want humans to be?

Fler avsnitt av Thinking On Paper