kuzu
Highly scalable, extremely fast, easy-to-use embeddable graph database
Decision gist · record as of 2026-08-14
Yes, if you need an embedded graph database for analytical workloads and can accept that the project is archived. The MIT license, zero runtime dependencies, and broad platform support make it low-friction to try. However, be aware that the maintainers are no longer actively developing new features—use it for stable, well-tested workloads, not for cutting-edge functionality or if you require ongoing vendor support.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires a compatible Python version and matching platform wheel; no external runtime dependencies, but the database engine is compiled C++.
- Medium install friction due to compiled wheels across multiple Python versions and platforms.
- The project is archived as of the latest release; prior versions remain usable but the maintainers are no longer actively developing new features.
License · maintenance · safety
MIT (permissive) — MIT license is permissive, allowing commercial and private use with minimal restrictions—suitable for most use cases without legal concern.
last release 2025-10-10 (308 days) · last repo commit 2025-10-10 · 4,026 stars · archived
0 known vulnerabilities (OSV.dev, 2026-08-14) · 436,393 downloads/mo, #6,676 on PyPI
Alternatives
Verify before relying
import kuzu
db = kuzu.Database(':memory:')
conn = db.get_connection()
result = conn.execute('MATCH (n) RETURN COUNT(n) AS count')
for row in result.get_as_df().itertuples():
print(row)- Whether the archived status affects long-term bug fixes or security patches for existing deployments
- Compatibility of pre-installed extensions (algo, fts, json, vector) with all use cases in v0.11.3
- Performance characteristics on specific graph sizes or query patterns
What it is and what it does
Kuzu is an embedded graph database designed to run inside your application without a separate server. It stores data as property graphs (nodes and edges with attributes) and queries them using Cypher, a declarative graph query language. The database is optimized for analytical workloads on large graphs, offering features like full-text search, vector indices for similarity queries, and ACID transactions. It uses columnar storage and specialized join algorithms to achieve fast query execution across multiple CPU cores.
The package provides a Python binding to the Kuzu engine. You embed it directly in your application, point it to a local data directory, and issue Cypher queries programmatically. As of version 0.11.3, four commonly used extensions (algo, fts, json, vector) are pre-installed; earlier versions or additional extensions require setting up a local extension server. The project is now archived, meaning the maintainers have moved to a new initiative, but existing code using prior releases continues to work without modification.
Use it for
- Build a recommendation engine by storing user-product relationships and querying graph patterns to find similar users or products.
- Analyze knowledge graphs or ontologies embedded in your application without deploying a separate database service.
- Perform full-text and vector similarity search on documents or embeddings stored as graph nodes.
- Run complex analytical queries on social networks, supply chains, or other highly connected datasets within a single process.
- Prototype graph-based features in a Python application before migrating to a production graph database.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you need an embedded graph database for analytical workloads and can accept that the project is archived.
The MIT license, zero runtime dependencies, and broad platform support make it low-friction to try. However, be aware that the maintainers are no longer actively developing new features—use it for stable, well-tested workloads, not for cutting-edge functionality or if you require ongoing vendor support.
Install
kuzu on PyPI
Before you install
Medium install friction due to compiled wheels across multiple Python versions and platforms. The project is archived as of the latest release; prior versions remain usable but the maintainers are no longer actively developing new features.
Requires a compatible Python version and matching platform wheel; no external runtime dependencies, but the database engine is compiled C++.
License in practice
MIT license is permissive, allowing commercial and private use with minimal restrictions—suitable for most use cases without legal concern.
Quickstart
import kuzu
db = kuzu.Database(':memory:')
conn = db.get_connection()
result = conn.execute('MATCH (n) RETURN COUNT(n) AS count')
for row in result.get_as_df().itertuples():
print(row)Verify before relying
- Whether the archived status affects long-term bug fixes or security patches for existing deployments
- Compatibility of pre-installed extensions (algo, fts, json, vector) with all use cases in v0.11.3
- Performance characteristics on specific graph sizes or query patterns
Package facts
| License | MIT permissive |
| Python support | Not specified |
| Install friction | Medium. Platform-specific wheel |
| Runtime dependencies | None |
| Maintenance | Abandoned 308 days since the last release |
| Last repo commit | repository archived |
| First released | |
| Downloads | 436,393 / month, #6,676 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: kuzu-0.11.3-cp310-cp310-macosx_11_0_arm64.whl; kuzu-0.11.3-cp310-cp310-macosx_11_0_x86_64.whl; kuzu-0.11.3-cp310-cp310-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl; kuzu-0.11.3-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl; kuzu-0.11.3-cp310-cp310-win_amd64.whl; kuzu-0.11.3-cp311-cp311-macosx_11_0_arm64.whl; kuzu-0.11.3-cp311-cp311-macosx_11_0_x86_64.whl; kuzu-0.11.3-cp311-cp311-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl; kuzu-0.11.3-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl; kuzu-0.11.3-cp311-cp311-win_amd64.whl; kuzu-0.11.3-cp312-cp312-macosx_11_0_arm64.whl; kuzu-0.11.3-cp312-cp312-macosx_11_0_x86_64.whl; kuzu-0.11.3-cp312-cp312-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl; kuzu-0.11.3-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl; kuzu-0.11.3-cp312-cp312-win_amd64.whl; kuzu-0.11.3-cp313-cp313-macosx_11_0_arm64.whl; kuzu-0.11.3-cp313-cp313-macosx_11_0_x86_64.whl; kuzu-0.11.3-cp313-cp313-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl; kuzu-0.11.3-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl; kuzu-0.11.3-cp313-cp313t-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “embedded graph database”
- kuzuKuzu is an embedded graph database that executes Cypher queries on…
- real-ladybugEmbeddable graph database with Cypher query support, full-text…
- owlready2Owlready2 is a Python module for ontology-oriented programming that…
Give your agent the search over MCP, or paste the wish link into any chat.
More Database packages
psycopg2-binary is a PostgreSQL database adapter for Python that implements the DB API 2.0 specification, enabling Python applications to connect to and query PostgreSQL databases with thread-safe concurrent operations.
Python client library for connecting to and executing commands against Redis key-value stores, supporting both synchronous and asynchronous operations.
Install it if your application needs to interact with Redis; the only prerequisite is a running Redis server instance.
YDB Python SDK is the official client library for connecting to and querying YDB databases from Python applications.
Install it if you need to connect Python applications to YDB databases.
Connects Python applications to Snowflake data warehouses using the DB API 2.0 specification, enabling SQL queries, data transfers, and warehouse operations.
sqlparse tokenizes SQL text into a tree of statements, clauses, and expressions, and provides functions to split scripts, format queries, and inspect parsed tokens without validating dialect or syntax.
Install it if you need to manipulate, format, or analyze SQL text programmatically.
Provides base adapter protocols and shared functionality that database adapters use to integrate with dbt-core, handling connections, dialect translation, relation caching, and core interface management.
See also real-ladybug · graph-retriever · FalkorDB · langchain-graph-retriever · nucliadb-utils · langchain-neo4j · neomodel · neo4j-graphrag · duckdb