SPONSORSHIP PROGRAM · OPEN CALL 2026

Quantum Flywheel

by BlueQubit

AI agents, cloud compute, quantum hardware access, and a quantum-native development environment — put into the hands of researchers with promising ideas and no budget to test them. Frontier AI has collapsed the path from hypothesis to working experiment from months to days. Compute access is now the bottleneck. Quantum Flywheel removes it.

Apply now

Registration opens September 14, 2026.

$150K

PROGRAM BUDGET · BLUEQUBIT + AWS COMPUTE + IBM QUANTUM COMPUTE

3

RESEARCH TRACKS

8–10wks

OPEN CALL FOR PROPOSALS

6mo

PROGRAM TERM

WHY NOW

Frontier AI research and coding assistants have removed the classic gates on a quantum result — building the simulation stack, tuning circuits, wrangling tooling. A researcher can now draft a full experimental pipeline in an afternoon.

That pipeline still stalls the moment it needs real QPU time and real classical horsepower behind it. Quantum Flywheel closes the gap: AI-accelerated ideation, plus AWS compute, plus BlueQubit's quantum R&D platform, which itself runs on AWS.

Intended outcome: publishable results and open artifacts — preprints, repositories, and reproducible benchmark data.

WHAT SPONSORED TEAMS GET

Quantum hardware access — QPU time on the latest IBM quantum computers, the core of the program.

Classical compute — Amazon EC2 GPU instances and large 16–120 core CPU machines for simulation, circuit compilation and optimization, and verification runs.

AI agents — frontier research and coding assistants to go from idea to pipeline in days.

BlueQubit platform — a quantum-native development environment with GPU-accelerated simulation, running on AWS.

THREE TRACKS — PROPOSALS COMBINING QUANTUM, AI, AND HEURISTIC METHODS

01

Quantum advantage on NISQ hardware

Experiments that push today's devices toward results classical methods cannot easily match.

02

Breaking quantum advantage claims

Adversarial classical simulation of published results — tensor networks, Pauli-path, and heuristic methods.

03

AI for new error correction codes

Using AI to discover, decode, and evaluate new QEC codes and decoders.

HOW IT WORKS

Open call

Proposals collected over an 8–10 week window.

Selection

The most promising submissions across the three tracks are funded.

Build & run

Teams get credits, QPU access, and the BlueQubit platform for the 6-month term.

Publish

Preprints, open repositories, and reproducible benchmark data.