Skip to simulation clusters
Time traveling companyExplore
Detailed compute blade with copper cold plates and braided liquid-cooling tubes

Simulation clusters

AI hardware for simulated time travel.

We co-design chips, racks, software, cooling, and manufacturing methods to compute multiple versions of our world with higher fidelity, measurable uncertainty, and reproducible results.

AI compute

One stack,
many possible worlds.

World simulation is not one model running once. It is a coordinated system that can hold a world state, populate it with many interacting agents, fork alternatives, and preserve the evidence behind every result.

Our research asks what changes when that workload shapes the complete machine—from silicon and interconnects to software, cooling, and the way a cluster is manufactured.

From chip to cluster

Designed as one instrument.

The system grows without changing its purpose: keep more of a simulated world active, connected, and inspectable at once.

A conceptual hardware family progressing from an accelerator chip to clustered compute racks

World accelerator

A purpose-shaped accelerator concept for agent inference, causal graphs, and long-horizon state transitions.

System architecture

Compute the branch.
Preserve the reason.

A hybrid stack coordinates classical supercomputing today while keeping a clear research boundary around quantum methods.

Cluster hardware showing a network switch, accelerator board, copper cold plate, compute tray, and three server racks

Evidence, agents and alternative decisions

World runtimeDefines the world and the experiment
OrchestrationSchedules runs, forks and checkpoints
Supercomputer fabricMoves state between compute and storage
Compute racksAI accelerators · CPU hosts · local memory
Shared storageEvidence · checkpoints · run history
QPU researchOptional task offload and returned samples

Power and liquid cooling support the compute racks. The quantum path is an exploratory interface for selected workloads.

Comparable futures, uncertainty and an auditable run record

Now viewing

World runtime

Compiles evidence, agents, environments, interventions, and evaluation rules into reproducible simulation runs.

Design intent
Many worlds / one experiment

Research direction. Components, interfaces, and performance remain exploratory rather than released product specifications.

From transistor to timeline

Co-design across every scale.

Each layer is optimized around the same job: explore consequences while keeping the path from evidence to outcome visible.

  1. Chips

    Purpose-shaped compute for repeated inference, interaction graphs, sampling, and state transitions.

  2. Racks

    Memory, interconnect, liquid cooling, and power delivery designed for long-running worlds.

  3. Software

    A runtime that compiles assumptions into branches and measures where their trajectories diverge.

  4. Manufacturing

    Modular assemblies and serviceable cooling paths intended to scale without hiding system behavior.

Four scientists examining branching simulated worlds in an advanced research laboratory

World forks

Run the decision more than once.

A simulation cluster can hold a shared starting world, introduce one controlled difference, and evolve many futures in parallel. The value is not a single prophecy. It is seeing which outcomes remain stable and which assumptions change everything.

Possible-world branching diagramA shared evidence baseline branches into resilient, accelerated, constrained, and disrupted possible futures.Evidence baselineResilientAcceleratedConstrainedDisrupted

Quantum research layer

Qubits where they help. Classical compute everywhere else.

Quantum processors are not general replacements for supercomputers. Our long-horizon research asks whether carefully selected QPU workloads could assist sampling, optimization, or search inside a much larger classical simulation system.

Qubits

Quantum information units whose useful behavior depends on carefully controlled superposition and entanglement.

Coherence

The limited interval in which fragile quantum states remain usable before noise overwhelms them.

Error correction

Methods that combine many physical qubits to protect a smaller number of logical qubits.

Research direction only. OpenX does not claim a deployed QPU, fault-tolerant quantum computer, or demonstrated quantum advantage.

A conceptual cryogenic quantum processor with gold, copper, and coaxial control hardware
concept quantum-control assembly

Research targets

Accuracy is measured,
not promised.

Simulation quality comes from validation, transparent uncertainty, and disciplined comparison with reality—not from claiming a perfect forecast.

Fidelity

Preserve the relationships that materially shape an outcome.

Uncertainty

Expose confidence, assumptions, and plausible alternatives.

Reproducibility

Replay a branch from the same evidence and configuration.

Validation

Continuously compare simulated behavior with observed reality.

Concept of a home hosting a dedicated GPU compute node

Distributed AI compute

A network beyond the data center.

Explore dedicated GPU nodes hosted in homes, offices, shops, warehouses, enterprise sites and suitable public facilities, coordinated to run independent simulation branches.

Explore distributed compute

Partners in compute

Build the machine for worlds that do not exist yet.

We want to work with researchers, infrastructure teams, manufacturers, and technical partners who think world simulation deserves its own compute stack.

Build with us