QBoltz
QBoltz builds a free-space optical computer that solves NP-hard combinatorial optimization problems—expressed as QUBO or Ising models—in under 10 milliseconds using a single light-interference pass. The room-temperature, battery-compatible system scales to millions of fully connected variables and is designed for edge deployments like drone swarms and robot fleets. Sampling is planned for 2027.
- Hardware
Funding
Founders
Founder details are not available yet.
Product
Problem
NP-hard combinatorial optimization problems—such as drone swarm re-planning, robot fleet re-tasking, and delivery network rerouting—are computationally intractable for conventional hardware, requiring millions of operations to approximate. Quantum computers are being developed to address these problems, but they demand cryogenic environments, shielding, and data-center-scale infrastructure that limit deployment. Real-time autonomy applications cannot wait for round trips to remote compute resources or tolerate the latency of existing approaches.
Solution
QBoltz computes using light interference instead of electrons. A single optical pass evaluates an entire candidate solution at once, doing in one operation what conventional hardware needs millions of operations to approximate. The free-space optical computer is packaged small enough to deploy anywhere and operates at room temperature with no cooling, vacuum, or shielding. It solves NP-hard combinatorial optimization problems expressed as QUBO or Ising models, scaling to millions of fully connected variables, with time-to-solution under 10 milliseconds. The system runs on tens of watts, making it battery-compatible for edge and autonomous deployments.
Target Audience
Primary customers are organizations developing real-time autonomy systems, including drone swarm operators, robotics fleets, and logistics networks that need sub-10-millisecond optimization decisions in the field.
Features
- Free-space optical computing architecture that performs whole-solution evaluation in a single optical pass
- Supports NP-hard combinatorial optimization expressed as QUBO or Ising models
- Scales to millions of fully connected variables
- Time to solution under 10 milliseconds
- Room-temperature operation with no cooling, vacuum, or shielding required
- Power consumption in the tens of watts range, battery-compatible
- Compact form factor deployable outside data centers