- API Type
- REST API with HTTP endpoints, plus WebSocket streaming for real-time data. Binary DBN encoding for performant zero-copy market data decoding.
- Authentication
- API Key authentication (32-character string). Keys managed through API keys portal. Support for environment variable passing (DATABENTO_API_KEY).
- Official SDKs
- Python client library (primary), C++ client, Rust client. All SDKs available on GitHub at github.com/databento/databento-python and support historical and live APIs.
- API Capabilities
- Live API: real-time subscriptions and intraday replay within 24 hours. Historical API: access to data older than 24 hours. Support for multiple data schemas (trades, order book, CBBO, BBO), multiple datasets, and various symbology types (raw_symbol, instrument_id, parent, continuous).
- Data Formats
- DBN binary format (primary), with conversion capabilities to CSV, JSON, and Parquet formats. Consistent schemas and record structures across live and historical APIs.
- Documentation
- Comprehensive API documentation with code examples, tutorials, and use-case guides at databento.com/docs. Examples organized by language (Python, C++, Rust), use case, and schema type.
- Streaming Capabilities
- Live client built with Python's asyncio module for easy integration into asynchronous applications. Support for synchronous and asynchronous iteration patterns. Callback-based event handling with error handlers and reconnection callbacks.
- Data Subscriptions
- Multiple subscriptions combinable into single session: trades, subsampled data (second, minute, hour, daily aggregates), definitions (expirations, settlement). Support for all symbols or specific instrument filters.
- Use Cases
- Live order book state retrieval, historical backtesting, portfolio valuation, technical indicator calculation (VWAP, RSI, Sharpe ratio), order book construction from MBO data, trade halt handling, data aggregation and resampling.