Vector Search & ANN Algorithms - Interactive Visualization

Vector search finds the semantically closest documents to a query by comparing embedding vectors. Exact search is accurate but O(n). HNSW builds a navigable small-world graph for O(log n) approximate search. IVF clusters vectors into buckets for fast coarse filtering. This demo animates the search traversal and shows recall vs latency tradeoffs.

• Exact search: O(n) exhaustive comparison, 100% recall, slowest at scale
• HNSW traversal: animated graph walk from entry point to approximate nearest neighbors
• IVF bucket search: coarse quantization prunes the search space before exact comparison
• Recall vs latency chart: tune efSearch and nprobe to see the tradeoff curve shift

Part of the EngineersOfAI Interactive 3D - free interactive visualizations covering every major concept in machine learning and AI engineering. Hover any element for a plain-English explanation. No code required.