Kuzu V0 136 !!better!! Official

Kùzu version (v0.13.6) is an update to the embedded, highly scalable property graph database designed for analytical workloads. This release continues Kùzu's focus on speed and massive graph processing using a columnar storage engine. Key Features & Updates in v0.13.6 According to official GitHub Release Notes Kùzu Documentation

Surrounding text or system
Any other numbers, device names, or software names nearby? kuzu v0 136

Migration Guide: Upgrading to v0.136

If you are currently on a previous version (e.g., v0.120 or v0.130), note the following breaking changes: Kùzu version (v0

The Kuzu team is working on several features and improvements, including: No distributed mode – You cannot cluster multiple

• No distributed mode – You cannot cluster multiple Kuzu instances. It is single-node, though multi-threaded.
• Graph algorithms library – Unlike Neo4j GDS, Kuzu does not have built-in PageRank, Louvain, or Label Propagation. You must implement these via Cypher iterations.
• Full-text indexing – Partial support via CONTAINS predicate, but no inverted indices for tokenized search.

2. Knowledge Graph Embedding

Data scientists building RAG systems (Retrieval-Augmented Generation) need to store entity relationships. The new LIST of STRUCT type allows you to attach vector embeddings directly to nodes as a list of floats, eliminating the need for a separate vector database.

• A known issue with data import has been identified and is being addressed in a future release.

By focusing on the embedded use case, Kuzu enables:

Query (new in v0.136: using LIST)

result = conn.execute("MATCH (a:Person) RETURN a.name, [ (a)-[:Knows]->(b) | b.name ] AS knows_list") print(result.get_as_data_frame())