LadybugDB is an embedded columnar graph database with a Cypher query engine — DuckDB-style ergonomics for connected data. Zero-config single file, vectorized execution, with Postgres interop, lakehouse interop, and a surrounding ecosystem for analytics, visualization, agent filesystems, and RDF/SHACL.
curl -fsSL https://install.ladybugdb.com | sh
· · ·
## Query Demo Datasets on Huggingface
lbug> ┌──────────┬─────────┬──────────┐ │ src.name │ r.name │ dst.name │ │ STRING │ STRING │ STRING │ ├──────────┼─────────┼──────────┤ │ user │ follows │ user │ │ user │ livesin │ city │ └──────────┴─────────┴──────────┘ (2 tuples) (3 columns) Time: 2.58ms (compiling), 0.38ms (executing)
lbug> ┌───────┬─────────┬───────┬─────────┬───────┬───────────┬──────────────┐ │ a.id │ a.name │ a.age │ b.since │ c.id │ c.name │ c.population │ │ INT32 │ STRING │ INT64 │ INT32 │ INT32 │ STRING │ INT64 │ ├───────┼─────────┼───────┼─────────┼───────┼───────────┼──────────────┤ │ 75 │ Noura │ 25 │ 0 │ 500 │ Guelph │ 75000 │ │ 100 │ Adam │ 30 │ 0 │ 700 │ Waterloo │ 150000 │ │ 250 │ Karissa │ 40 │ 0 │ 700 │ Waterloo │ 150000 │ │ 300 │ Zhang │ 50 │ 0 │ 600 │ Kitchener │ 200000 │ └───────┴─────────┴───────┴─────────┴───────┴───────────┴──────────────┘ (4 tuples) (7 columns) Time: 1.33ms (compiling), 2.07ms (executing)
lbug> ┌───────┬─────────┬───────┬─────────┬───────┬─────────┬───────┐ │ a.id │ a.name │ a.age │ b.since │ c.id │ c.name │ c.age │ │ INT32 │ STRING │ INT64 │ INT32 │ INT32 │ STRING │ INT64 │ ├───────┼─────────┼───────┼─────────┼───────┼─────────┼───────┤ │ 100 │ Adam │ 30 │ 2020 │ 300 │ Zhang │ 50 │ │ 100 │ Adam │ 30 │ 2020 │ 250 │ Karissa │ 40 │ │ 250 │ Karissa │ 40 │ 2021 │ 300 │ Zhang │ 50 │ │ 300 │ Zhang │ 50 │ 2022 │ 75 │ Noura │ 25 │ └───────┴─────────┴───────┴─────────┴───────┴─────────┴───────┘ (4 tuples) (7 columns) Time: 1.07ms (compiling), 2.00ms (executing)
## Ecosystem
One storage engine, several ways to reach it. Everything below reads and writes the same graph.
Columnar graph analytics that works directly on LadybugDB, Parquet, and Arrow — no export step. PageRank, components, community detection, paths.
Lean, Rust-backed graph visualizer for exploring LadybugDB — built for large result sets, not slideware.
Rich extension system: adbc, postgres, duckdb, iceberg, delta, httpfs, json, fts, vector, algo, neo4j and more. INSTALL adbc; and query.
Run embedded in-process, or serve it: ADBC for Arrow-native clients and Bolt for Neo4j-compatible drivers.
Query Snowflake and Databricks lakehouses live via ADBC — attach the warehouse and join it against your local graph.
Runs the graph engine inside PostgreSQL. Cypher over your Postgres tables, native execution under the hood.
A filesystem view of the graph for agents that prefer writing text files: browse with ls, read with cat, mutate by rewriting files.
OWL ontologies, SHACL shapes, and RDF data stored as a Cypher property graph — validation on load, DDL constraining later writes.