surrealdb/surrealdb

Features

• Multi-Model Databases - Stores and queries document, graph, relational, and vector data within a single ACID-compliant engine.
• ACID Transactional Cores - Groups multiple operations into atomic, isolated units to ensure data integrity and consistency across complex distributed database workloads.
• Database Engines - Maintains data integrity using ACID-compliant transactions, multi-tenancy isolation, and flexible schema modes.
• Declarative Query Languages - Provides a unified declarative syntax for CRUD, graph traversal, and vector search.
• Access Control Systems - Enforces granular, record-level permissions and multi-tenant isolation directly at the data layer.
• Agent Memory Storage - Keeps agent memory inside the database to eliminate middleware layers while ensuring consistent permissions.
• Vector Databases - Integrates vector search, graph traversal, and machine learning model execution into the query layer.
• Database Client Libraries - Provides a strongly-typed client interface for consistent communication with the database instance.
• Distributed Transaction Processing - Processes write transactions through a quorum consensus mechanism to ensure data durability across multiple nodes.
• Graph Querying - Supports complex graph traversal and nested object projection for connected data.
• Multi-Model Vector Storage - Keeps vector embeddings alongside structured data and graph relationships within a single database to simplify data management.
• Query Languages - Manipulates data using a multi-model language that supports standard CRUD operations and graph-based relationship traversal.
• Access Control Policies - Evaluates granular security and access control rules directly against individual data records during query execution.
• Agent Context Management - Provides a unified memory layer that maintains consistent state and permissions across multiple database models for intelligent agents.
• Agent Memory Systems - Maintains persistent, ACID-consistent memory and knowledge graphs for intelligent agents.
• Atomic Agent Contexts - Handles memory, knowledge graphs, and structured data within a single transaction boundary to ensure data consistency.
• In-Database Model Execution - Runs trained machine learning models directly within database queries to perform real-time predictions on stored data.
• Browser Databases - Provides high-performance data storage directly within the browser environment.
• Consensus Protocols - Processes write transactions through a distributed agreement mechanism that ensures data durability and consistency.
• Database Clients - Instantiates database clients to handle connections using multiple protocols.
• Database Execution Engines - Performs document storage, search, vector processing, and graph querying natively within a single unified engine.
• Distributed Databases - Operates across embedded, edge, and cloud environments using a consistent binary and API.
• Event-Driven Subscription Systems - Pushes real-time data updates directly to connected clients via live queries to eliminate the need for external message brokers.
• Indexing and Search - Executes complex data lookups using unique, compound, and array indexes alongside full-text and vector indexing.
• Real-Time Data Streaming - Powers reactive, real-time experiences using built-in subscriptions, event triggers, and streaming updates.
• Relationship Management - Connects related data records and performs vector similarity searches natively within a single database.
• Transaction Management - Manages query execution flow and manual transactions to ensure data consistency.
• Model Context Protocol Integrations - Links AI models directly to database services using standard protocol tool calls to avoid custom integration code.
• Multi-Agent Coordination - Synchronizes multiple agents using shared memory and event-driven handoffs to maintain consistent state across complex workflows.
• Database Management Tools - Scalable document-graph database for real-time web applications.
• Database Systems - Scalable, distributed document-graph database.
• Databases - Listed in the “Databases” section of the Awesome Mac awesome list.
• Vector Databases - Scalable multi-model database optimized for time-series data.
• Data Change Subscriptions - Pushes data changes to connected clients instantly using live queries.
• Database Administration - Organizes database resources by managing schema elements and access grants.
• Edge Databases - Deploys a consistent database engine across environments ranging from edge devices to cloud clusters.
• Graph Databases - Handles continuously updated graph data and multi-model workloads at scale.
• Hybrid Search - Merges vector similarity searches with graph traversal and structured data filters in a single query.
• In-Memory Databases - Manages application state using a high-speed embedded in-memory engine.
• Reactive Databases - Pushes real-time updates to connected clients through live queries and event triggers.
• Schema Definition Tools - Creates type-safe table references to ensure language compatibility and consistent data access patterns.
• Storage-Compute Architectures - Decouples the database processing layer from underlying storage nodes to enable independent scaling.
• Distributed Systems - Distributes database instances across industrial environments from edge devices to cloud infrastructure.
• Fault Tolerance - Restores failed nodes by retrieving state from object storage and replaying transaction logs.
• Data Security Frameworks - Enforces granular, record-level permissions and multi-tenant isolation across all stored data.
• Agent Memory Architectures - Structures agent data across working, semantic, and episodic storage using graph relationships and temporal awareness to support specialized recall.
• Database Assistants - Generates database queries, designs schemas, and troubleshoots errors using an integrated assistant that understands workspace context.
• Multi-Axis Memory Querying - Searches memory across semantic, relational, and temporal axes using a unified substrate that fuses signals into a single ranking.
• Unified Data Querying - Retrieves diverse data types including documents, graphs, and vectors through a single query language and protocol interface.
• Connection Establishment - Connects to database instances using network or embedded protocols with authentication.
• Data Modeling - Represents diverse data structures using native support for arrays, objects, datetimes, and geometry types.
• Key-Value Stores - Stores information in an embedded database using a reliable key-value interface.
• Remote Query Execution Protocols - Executes multiple complex database queries within a single network request using persistent websocket connections.
• Versioned Storage - Maintains historical data snapshots using an immutable tree structure.
• Single-Binary Distributions - Deploys a single-binary database engine across diverse environments while maintaining a consistent interface.
• Object Storage Integration - Saves transactional data directly in cloud object storage to enable stateless, elastic compute.
• WebAssembly - Compiles the core database engine into portable bytecode to enable high-performance execution within browser and edge environments.
• Access Governance - Restricts extension execution using custom permission expressions to ensure secure multi-tenant access.
• Reactive Data Streams - Streams database updates directly to clients to power live, event-driven user experiences.
• Agent Memory Categorization - Organizes memory into distinct categories like episodic, identity, and knowledge with specific schemas and retrieval weights.
• Agent Memory Maintenance - Runs background processes that autonomously discover connections between entities and consolidate fragmented knowledge for intelligent agent applications.
• Memory Provenance Tracking - Records fact provenance as first-class data using independent system clocks to maintain an auditable history of information.
• Semantic Substrates - Creates a multi-model substrate that anchors canonical meaning and governs access across an organization.
• Cloud Database Provisioning - Enables automated deployment and configuration of cloud-hosted database instances with specific performance and scaling requirements.
• Connection Pooling - Establishes persistent connections to database instances for efficient data access.
• Data Insertion Interfaces - Maps application objects to database records while supporting auto-generated and custom identifiers for data integrity.
• Data Retrieval Interfaces - Fetches specific records using unique identifiers or custom query statements with bound parameters.
• Schema Management - Structures domain ontologies directly within the database schema using tables and typed relations.
• Semantic Search - Improves product discovery by matching items using semantic search and AI embeddings.
• Managed Database Services - Operates database clusters in the cloud with automated infrastructure and independent compute scaling.
• Execution Isolation - Executes custom modules within isolated memory sandboxes to maintain system stability and security.
• Sandboxing Environments - Executes custom user-defined modules within restricted memory environments to prevent unauthorized system access.
• Schema Migration Tools - Tracks and manages schema changes to ensure compatibility during data evolution.

Star history