NVIDIA aren't content with dominating AI, they have their sights set on quantum computing.

They have partnered with Infleqtion to bridge the gap between traditional GPU-powered supercomputers and quantum.

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.

And they've already achieved a four-microsecond round-trip latency. That's fast as lightning, to you and me. And it means 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.

Pranav Gokhale, CTO of Infleqtion, and Sam Stanwyck, Group Product Manager for Quantum Computing at Nvidia, Think On Paper.

--

• ⁠⁠Listen to every podcast⁠⁠

• Follow us on⁠ ⁠Instagram⁠⁠

• Follow us on⁠ ⁠X⁠⁠

• Follow Mark on⁠ ⁠LinkedIn⁠⁠

• Follow Jeremy on⁠ ⁠LinkedIn⁠⁠

• Read our⁠ ⁠Substack⁠⁠

• Email: ⁠[email protected]⁠

--

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?