HelixDB is a graph database for representing entities and their relationships.
It combines graph, vector, and full-text search in a single database.
All data forms one labeled property graph: nodes and directed edges can both contain typed properties.
The graph remains the source of truth; search indexes are access paths over properties rather than separate data stores.
It removes synchronization between graph, vector, and search systems, reducing architectural complexity.
Nodes and relationships are both searchable, including semantic search, text search, and property filtering.
ACID transactions span graph, vector, and textual data, providing consistent reads and writes.
Object storage holds the authoritative data, while memory and NVMe storage act as caches.
Storage scales independently from computation, allowing datasets to exceed the capacity of an individual server.
It supports ontologies and AI applications, including connected company knowledge, enterprise search, agent memory, and Helix Foundry’s unified view of organizational data.
Do you like what you are reading? Subscribe to receive updates.
Unsubscribe anytime