Embedding Atlas
Embedding Atlas is an MIT-licensed toolkit for working with large text or image embeddings. It provides interactive, in-browser visualizations with cross-filtering and metadata search, and ships as both a Python package (CLI tool, Jupyter widget, Streamlit component) and an npm UI component library. It computes embeddings and 2D projections locally (data stays on-device), and includes WebAssembly implementations of UMAP, approximate nearest-neighbor search, and density-based clustering. Integrations include SentenceTransformers for embedding generation and optional loading of datasets from Hugging Face.
No discussions yet
Be the first to start a discussion about Embedding Atlas
Developer
Apple publishes select open-source tools and research artifacts, including Embedding Atlas for exploring and visualizing embeddings.
Pricing and Plans
| Plan | Price | Features |
|---|---|---|
| Open Source (MIT) | Contact us |
|
System Requirements
Operating System
Modern web browser, Windows/macOS/Linux with Python 3.10+ (for CLI/Jupyter/Streamlit)
Memory (RAM)
8GB recommended for medium/large datasets
Processor
Modern multi-core CPU; WebAssembly acceleration in browser
Disk Space
Varies by dataset; embeddings and projections stored locally
AI Capabilities
Embedding visualization and exploration
Cross-filtering and semantic search over metadata
Dimensionality reduction with UMAP (WebAssembly)
Approximate nearest-neighbor search (HNSW, NN-Descent)
Density-based clustering and auto-label hints
Local, on-device computation and privacy
CLI-driven embedding/projection workflows
Notebook and app embedding via Jupyter/Streamlit components