cyborgdb-py and cyborgdb-js
These are ecosystem siblings—language-specific client SDKs for the same CyborgDB backend service, designed to be used independently depending on whether developers are building in Python or JavaScript/TypeScript environments.
About cyborgdb-py
cyborginc/cyborgdb-py
Python Client SDK for CyborgDB: The Confidential Vector Database
Supports encrypted vector ingestion, similarity search, and metadata filtering with client-side key management—all operations remain encrypted end-to-end. Built on a zero-trust architecture that delegates computation to a separate `cyborgdb-service` backend while keeping confidential data secure on the client. Integrates with standard Python ML workflows and supports GPU-accelerated indexing via CUDA with configurable index types like IVFFlat and IVFPQ.
About cyborgdb-js
cyborginc/cyborgdb-js
JavaScript & TypeScript Client SDK for CyborgDB: The Confidential Vector Database
Supports encrypted vector ingestion, search, and retrieval with client-side key management and zero-trust architecture. Communicates with a separately-deployed Python service via HTTP, enabling batch operations and metadata filtering while maintaining end-to-end encryption of embeddings. Full TypeScript type safety with configurable index types (IVFFlat, IVFPQ) for production vector workloads.
Scores updated daily from GitHub, PyPI, and npm data. How scores work