Small. Fast.
Lucky.
Ladybugs eat the pests and leave the garden alone. They also have excellent scent-based memory that lets them remember migration paths and trails left by their ancestors. LadybugDB does the same to your stack: one small embedded graph engine that removes servers, sidecars, and ops — and leaves the data.
zero-ops
vectorized
anywhere
embedded · cypher · MIT-licensed — see the comparisons
## By The Numbers
Knowledge-graph and context-graph storage for your agents — the scent trails your stack remembers.
Single-file embedded footprint: sqlite (< 1MB), DuckDB (~60MB), LadybugDB (~20MB).
220+ graph algorithms from icebug — PageRank, components, community detection, paths.
800k downloads a month across SDKs and extensions.
190 projects building on LadybugDB.
ADBC and Arrow Flight are significantly more efficient over the wire when run as a server.