GDS 2.20.0 release

Release Date: 31 July 2025

GDS 2.20 is compatible with Neo4j 2025.07. For GDS compatibility with previous releases, please use GDS Compatibility Table.

New Features

Add support for the new labelFilter parameter of the gds.triangleCount.* procedures which allow users to specify the labels of the nodes in the triangles counted by the algorithm.

Bug fixes

Fix a bug where trying to create graphs through native projection with incorrect parameters would allocate memory in the memory guard.

Graph Data Science 2.19.0

New features

• Added memory estimation for harmonic centrality algorithm.
• Added memory estimation for closeness centrality algorithm.
• Added memory estimation for all pairs shortest paths algorithm (stream mode).

Bug fixes

Improvements

• Promote all pairs shortest paths algorithm (stream mode) to prod tier.

Graph Data Science 2.13.4

neo4j-graph-data-science-2.13.4

Breaking changes

New features

Bug fixes

• Fix a bug with Node2Vec that could impact randomness generation.

Improvements

Other changes

Graph Data Science 2.16.0

Neo4j Graph Data Science Plugin changelog 2.16

Bug fixes

• Avoid hanging tasks for gds.graph.project (Cypher Projection).

Graph Data Science 2.15.0

New features

• gds.articulationPoints.stream can now return for each articulation point a map resultingComponents tracking the number of components, as well as their minimum/maximum size that are formed following the articulation point's removal.

Bug fixes

• Fixed a bug where algorithms would keep tasks around on failure. Affected algorithms:
  • gds.beta.pipeline.linkPrediction.train - Could not start another training on the same graph afterwards.
  • gds.dag.longestPath
  • gds.randomWalk
  • gds.shortestPath.yens
  • gds.allShortestPaths.delta

Improvements

• Native projections will now reject relationships which point to or from nodes that were not scanned during the node projection.
  This situation can happen when there are simultaneous modifications to the DBMS during projection.
• In gds.fastRP.{mutate,write,stream}, the initial random node vectors use a degree-scaling which is uniform over the entire vector, whereas previously a different scaling was applied to the property-aware part of the vector.
  This makes the relationship between propertyRatio and the influence of properties more intuitive, also when normalizationStrength is non-zero.

Graph Data Science 2.14.0

neo4j-graph-data-science-2.14.0

New features

• gds.bridges.stream can now return for each bridge a list with the sizes of the resulting components after the bridge is removed by the remainingSizes yields parameter.
• New write mode procedures for hashGNN:
  • gds.hashgnn.write
  • gds.hashgnn.write.estimate

Bug fixes

• Fixed a bug where gds.graph.sample.rwr and gds.graph.sample.cnarw would fail on very small graphs.
• Fix a bug where gds.spanningTree would return zero cost per relationship instead of one for unspecified relationshipWeightProperty.

2.13.2

neo4j-graph-data-science-2.13.2

Bug fixes

• GDS 2.13.1 was not compatible with Neo4j Community Edition 5.26, only with Neo4j Enterprise Edition.
  This is fixed in GDS 2.13.2, and it now works on both editions of Neo4j
• Fixed Maven dependency on neo4j which had an incorrect -SNAPSHOT suffix.

2.13.1

neo4j-graph-data-science-2.13.1

New features

• New algorithm for the Prize Collecting Steiner Tree. Supported modes:
  • gds.prizeSteinerTree.stream
  • gds.prizeSteinerTree.stream.estimate
  • gds.prizeSteinerTree.mutate
  • gds.prizeSteinerTree.mutate.estimate
  • gds.prizeSteinerTree.write
  • gds.prizeSteinerTree.write.estimate
  • gds.prizeSteinerTree.stats
  • gds.prizeSteinerTree.stats.estimate
• New procedures for tracking memory footprint:
  • gds.memory.list
  • gds.memory.summary

Bug fixes

• Fixed a bug where Cypher projections RETURN gds.graph.project(...) would leave not cleanup its progress tasks Graph aggregation in case of failure.
• Fixed an issue when gds.systemMonitor reports availableCpuCoresNotRequested incorrectly.
• Fixed CVE-2024-47535 in the netty dependency, affecting Windows users.

2.12.0

neo4j-graph-data-science-2.12.0

Bug fixes

• Fixed a bug to make algorithms once again respect the sudo flag, allowing then to bypass the memory guard.

Improvements

• Added support for stopping GraphSage and sampling algorithms when a termination signal is received.

2.11.0

• Fixed a bug which prevented projecting graphs from Neo4j with float-array node properties