Classical Algorithms
The foundations of computer science
Binary search, sorting, graph traversal, and shortest paths — visualized step by step with full code synchronization.
Interactive dissection of deep learning, generative AI, and quantum algorithms
Free & open source · No account required · MIT Licensed
From sorting algorithms to self-attention — explore every domain with the same interactive, step-by-step experience.
The foundations of computer science
Binary search, sorting, graph traversal, and shortest paths — visualized step by step with full code synchronization.
Neural networks from neuron to network
Watch signals propagate through neurons, see backpropagation compute gradients, and understand convolution at the pixel level.
Transformers and attention, demystified
See how text becomes tokens, tokens become embeddings, and attention computes which words matter to each other.
Qubits, gates, and superposition
Bloch spheres, quantum gates, and algorithms like Grover's search — visualized in ways that build real intuition.
Every feature is designed to help you build genuine intuition, not just watch animations.
Play, pause, step forward, step backward, and scrub through any algorithm at your own pace.
Every step highlights the exact line of code being executed. See the connection between logic and visualization.
Each step includes a human-readable explanation of what just happened and why it matters.
Modify the input data and parameters to see how the algorithm behaves differently. Real understanding comes from experimentation.
Every visualization state has a unique URL. Share the exact step you're looking at with anyone.
pip install eigenvue — use the same visualizations in Jupyter notebooks and Python scripts.
Install the Python package and use the same visualizations in Jupyter notebooks, scripts, and research workflows.
pip install eigenvuenpm install eigenvue1import eigenvue23# List all available algorithms4eigenvue.list()56# Visualize self-attention interactively7eigenvue.show("self-attention")89# Use in Jupyter notebooks10eigenvue.jupyter("transformer-block")
No account needed. No installation required. Pick an algorithm and start learning.