OPEN RESEARCH TOOLKIT

See what your quantum program is doing.

Q-VTK brings together visualization techniques for every layer of quantum computing: quantum states, circuits, noisy hardware and quantum neural networks.

Explore the modules ASPLOS 2026 tutorial
3-qubit measurement · noise spread per basis state illustrative
0.6 0.4 0.2 0 |000⟩ |001⟩ |010⟩ |011⟩ |100⟩ |101⟩ |110⟩ |111⟩ widest spread: flag as noisy
The kind of question Q-VTK answers: which output states can you trust, and which ones does the hardware distort?

Four layers, one toolkit

Quantum programs are hard to reason about because the things that matter cannot be printed: amplitudes, entanglement, gate effects, device noise and trained parameters. Q-VTK gives each of these a visual form.

|ψ⟩

Quantum states

Amplitudes, superposition and entanglement for one and two qubits.

VENUS

U(θ)

Quantum circuits

How each gate changes the state, step by step through the circuit.

QuantumEyes

ε ≠ 0

Noisy hardware

Device noise, compiled-circuit fidelity and which output states the noise distorts.

VACSEN · Basis-State Views · NoiseLens

QNN

Quantum machine learning

Data encoding, ansatz training and measurement inside quantum neural networks.

VIOLET · Encoder Explorer

The modules

Each module answers one kind of question. The tag on each card names the layer it serves.

VENUS

STATES

A 2D geometric view of quantum states that shows entanglement and superposition.

The Bloch sphere cannot show entanglement and can hide details through 3D occlusion. VENUS maps quantum amplitudes to 2D shapes grounded in the math of quantum computing, for single qubits and entangled pairs, and uses coordinated semicircles to show the probability distribution.

QuantumEyes

CIRCUITS

Explains what a quantum circuit does, gate by gate.

At the global level, a Probability Summary View, a State Evolution View and a Gate Explanation View trace how gates change the quantum state. At the local level, the Dandelion Chart shows how amplitudes determine the probability of each basis state.

VACSEN

NOISE

Noise-aware choice of quantum computer and compiled circuit.

Linked views track each machine's performance metrics over time, filter compiled circuits to the most promising ones, and compare qubit and gate usage. Fidelity and probability-distribution views show the expected result quality, so users can pick the setup with the highest fidelity instead of trial and error.

Basis-State Views

NOISE

Compares output distributions across quantum computers and simulators.

For the circuits of a quantum machine learning application, this module plots the distribution of every basis state on several quantum computers and on noise-free simulators, and draws a heat map of the KL distance between states. Users see at a glance where real hardware departs from the ideal result.

NoiseLens

NOISE

Separates noisy from less noisy basis states automatically.

A four-step pipeline: KL divergence between every pair of basis states, a 2D map by multidimensional scaling, k-means clustering, and violin plots that flag high-variance clusters as noisy. The result is far less cluttered than functional box plots.

Source code on GitHub

VIOLET

QML

Opens the black box of quantum neural networks.

Coordinated views explain the three quantum-specific parts of a QNN: how input data is encoded into quantum states, how the ansatz evolves during training, and what the final measurements mean. It includes new visual designs such as an augmented heat map.

Encoder Explorer

QML

Helps choose a data encoder for a QNN before training.

Picking a quantum data encoder is usually trial and error, because encoded states are hard to evaluate before training and there are few intuitive ways to compare encoders. This module visualizes the encoded quantum states so users can compare encoders and make an informed choice up front.

Use it across the whole workflow

The modules follow a quantum program from design to results. Start wherever your question is.

  1. Step 1 Understand the states VENUS
  2. Step 2 Debug the circuit QuantumEyes
  3. Step 3 Design the QNN Encoder Explorer · VIOLET
  4. Step 4 Pick the hardware VACSEN
  5. Step 5 Diagnose the results Basis-State Views · NoiseLens

TUTORIAL · ASPLOS 2026

Learn Q-VTK hands-on

We presented a tutorial on Q-VTK at the ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2026). It introduced the toolkit to researchers in computer architecture and systems.

Tutorial materials Slides: ASPLOS 2026 Q-VTK tutorial slides