GraphForge vs. PostgreSQL + Apache AGE
When is a graph extension preferable to a dedicated graph engine?
Apache AGE adds openCypher to PostgreSQL. Graphs are stored as PostgreSQL tables, one per label, and queried through a cypher() SQL function, so graph data shares transactions, backups, and access control with relational data. Its built-in analytics are limited to unweighted shortest paths. GraphForge is a dedicated embedded graph engine with 94 algorithms, which makes this a trade-off between operational consolidation and analytical depth.
| GraphForge | PostgreSQL + Apache AGE | |
|---|---|---|
| Model | Embedded property graph | Relational + graph extension |
| Version | 0.5.2 stable · 0.6.0-rc.52 | 1.8.0 (PG18 branch) (2026-07-09)18 |
| License | Apache-2.0 (since 0.5.0) | Apache-2.0 (AGE); PostgreSQL License (PostgreSQL) |
| Status | Actively maintained2 | Actively maintained18 |
On this page
Should I choose PostgreSQL + Apache AGE or GraphForge?
PostgreSQL + AGE to keep graph data inside Postgres; GraphForge when analysis needs real graph algorithms.
Choose GraphForge when
- You need weighted paths, centrality, community detection, or embeddings, not just BFS shortest paths.
- The analysis runs in a notebook or application and should not depend on a database server.
- You want results as Arrow tables and investigation history kept with the graph.
Choose PostgreSQL + Apache AGE when
- Graph data must live in the same PostgreSQL transactions, backups, and permissions as relational data.
- Your team already operates PostgreSQL, including managed services such as Azure Database for PostgreSQL.
- Queries mix SQL joins with Cypher patterns, and the graph analytics needed are light.
What can each one do?
23 of 26 capabilities differ. For 13, PostgreSQL + Apache AGE's docs did not confirm support, so they are marked Not established rather than No.
Each capability has a status and a source. “Not established” means we could not confirm the feature from PostgreSQL + Apache AGE's primary documentation. It does not mean PostgreSQL + Apache AGE lacks it.
| Capability | GraphForge | PostgreSQL + Apache AGE |
|---|---|---|
| Storage | ||
| Property graph model | Shipped Labeled property graph with typed properties. 1 | Extension AGE extension adds labeled vertices/edges with agtype properties, queried with openCypher. 1719 |
| Document model | Not offered No document-store model; properties accept nested, heterogeneous values. 1 | Yes PostgreSQL json/jsonb types; AGE 1.8.0 adds agtype<->jsonb bidirectional casts. 18 |
| Relational tables | Not offered Not a relational database; results arrive as Arrow tables. 1 | Yes Full PostgreSQL SQL; Cypher results can be joined with tables in SQL queries and CTEs. 1721 |
| Key-value access | Not offered No key-value API; data is a property graph. 1 | Not established No dedicated key-value model documented by AGE; jsonb/tables can emulate it. |
| Columnar storage | Shipped Durable projects store graph data as Arrow and Parquet. 4 | Not established Not documented by AGE or PostgreSQL core. |
| Vector search | Shipped Exact cosine retrieval over embeddings you supply; no ANN index. 5 | Extension Via separate pgvector PostgreSQL extension (third-party), not part of AGE. 24 |
| Full-text search | Shipped Tantivy-backed BM25; hybrid text and vector retrieval through find(). 5 | Yes PostgreSQL core full-text search (tsvector/tsquery) on relational data; not exposed in Cypher. 23 |
| Geospatial | Not offered No spatial types or indexes are documented. | Not established No AGE spatial support documented; PostGIS on relational tables is separate. |
| Execution | ||
| Transactional workloads (OLTP) | Partial Local analytical scope; durable atomic writes with queued and optimistic writer modes. 6 | Yes Graph writes run inside PostgreSQL transactions. 22 |
| Analytical workloads (OLAP) | Shipped Analytical verbs and Cypher over a local columnar project. 7 | Not established No dedicated analytical graph engine documented. |
| Graph traversal | Shipped Cypher MATCH patterns plus BFS, DFS, Dijkstra, A*, and reachability. 8 | Yes openCypher MATCH pattern traversal. 20 |
| Variable-length paths | Shipped Bounded and unbounded patterns and shortestPath(). 8 | Yes MATCH supports variable-length edges such as *2, *3..5, open-ended ranges. 20 |
| Graph algorithms | Shipped 94 native algorithms behind analysis verbs, not Cypher CALL procedures. 37 | Partial 1.8.0 adds unweighted BFS shortest_path/all_shortest_paths only; no algorithm library. 2518 |
| Incremental computation | Not offered Algorithms recompute on request; no incremental mode is documented. 7 | Not established Not documented. |
| Deployment | ||
| Embedded / in-process | Shipped In-process library; no server, daemon, or connection string. 9 | Not established Runs inside a PostgreSQL server process; no embedded mode documented. |
| Server mode | Planned Optional remote peer extension planned after 0.6.0. 11 | Yes Installed into a PostgreSQL server with CREATE EXTENSION; Docker image provided. 17 |
| WebAssembly | Not planned A browser WebAssembly engine was declined in ADR 0027. 12 | Not established Not documented. |
| Distributed cluster | Not planned Distributed clusters and HA replication are out of scope. 9 | Not established No sharded/distributed graph deployment documented by AGE. |
| Managed cloud | Not planned No hosted compute. The graphforge.sh Hub distributes repositories; execution stays local. 13 | Partial No AGE-run cloud; offered as an extension on Azure Database for PostgreSQL flexible server. 22 |
| Investigation | ||
| Branching | Experimental Research versions, branches, slices, proposals, and forks ship in 0.6.0 release candidates. 14 | Not established Not documented. |
| Snapshots / time travel | Shipped Named checkpoints you can diff and revert, plus transaction-time snapshots. 3 | Not established Not documented. |
| Immutable history | Shipped Knowledge ledger records immutable assertions and evidence lineage. 16 | Not established Not documented. |
| Provenance | Shipped Provenance and knowledge ledger with evidence lineage. 16 | Not established Not documented. |
| Competing hypotheses | Shipped Immutable hypothesis groups with membership and selection events. 3 | Not established |
| Operations | ||
| ACID transactions | Partial Acknowledged-durable atomic publication; docs do not claim general ACID or serializable isolation. 6 | Yes Uses PostgreSQL's ACID-compliant transactional system for graph queries. 22 |
| Replication | Not planned HA replication is out of scope. 9 | Yes Inherited from PostgreSQL physical/streaming replication; not AGE-specific. 26 |
Showing 26 of 26 capabilities.
How is each one built?
PostgreSQL + Apache AGE: relational + graph extension. GraphForge: an embedded property graph library in Rust, persisted as Parquet.
GraphForge
GraphForge is a Rust library that runs inside your process: no daemon, server, or connection string. Projects persist to a local directory as Arrow and Parquet data plus JSON metadata, and every query and analysis verb returns an Apache Arrow table. Algorithms run in the same native engine as Cypher and are called as API verbs (rank, cluster, paths, analyze, similar), not as Cypher procedures.
PostgreSQL + Apache AGE
AGE is a C extension running inside the PostgreSQL server. Each graph gets its own schema and each label its own table, with metadata in ag_catalog. Cypher runs through cypher(graph, query, params), which returns records that need a column definition list. Release branches track PostgreSQL 11 through 18.
AGE adds graph queries to an existing relational system. GraphForge is a separate engine built only for graph analysis. Many teams keep the system of record in PostgreSQL and analyze an exported graph.
What the documentation says about PostgreSQL + Apache AGE
- Apache AGE graduated from the Apache Incubator and was announced as an ASF top-level project on 2022-06-08.27
- AGE is an acronym for A Graph Extension and is inspired by Bitnine's AgensGraph, a multi-model fork of PostgreSQL.17
- Cypher is invoked through the SQL function cypher(graph_name, query_string, parameters), which returns SETOF records requiring a column definition list.21
- Each graph gets its own PostgreSQL namespace, and each vertex or edge label is backed by its own table in that namespace.19
- Graph metadata lives in the ag_catalog schema tables ag_graph and ag_label.19
- The README states support for PostgreSQL 11 through 18, with separate release branches per PostgreSQL major version.17
- AGE 1.8.0 release notes add MERGE ON CREATE SET / ON MATCH SET, reduce(), create_subgraph(), and agtype<->jsonb casts.18
- Official drivers for Go, Java (JDBC), Node.js and Python live in the AGE repository; Rust and .NET drivers are community-driven.17
Which one is faster?
Not measured yet. No head-to-head run has been published, so this page shows no performance chart.
Charts appear only after a reproducible run exists: pinned versions, disclosed hardware, generated data, raw results, and scripts.
No comparative results yet
The suites below apply to GraphForge and PostgreSQL + Apache AGE. Each will be reported separately per benchmark class, never merged into one leaderboard.
Ingestion and writes
| Suite | Workloads | Metrics | Status |
|---|---|---|---|
| INGEST | CSV, JSONL, Parquet, graph edge lists | Records/s, MB/s, total elapsed | Not run |
Query, analytics, and system
| Suite | Workloads | Metrics | Status |
|---|---|---|---|
| LOOKUP | ID lookup, indexed property filter | p50/p95/p99 latency | Not run |
| TRAVERSE | BFS, k-hop, variable-length path | Latency, visited edges/s | Not run |
| JOIN | Multi-relation pattern matching | Latency, rows/s | Not run |
| ALGO | PageRank, WCC, SCC, Louvain, SSSP | Time, memory, correctness | Not run |
| CONCUR | Mixed readers and writers | Throughput, tail latency | Not run |
Which graph algorithms ship?
GraphForge: 94 native algorithms. PostgreSQL + Apache AGE: 3 named in its docs, delivered as extension.
Algorithms are listed by name, as documented. Counts of extension modules or third-party libraries are not presented as native coverage.
GraphForge
94 Rust algorithms run in the engine behind five analysis verbs (rank, cluster, paths, analyze, similar) and return Arrow tables.37
A selection of the catalog:
- PageRank
- ArticleRank
- Betweenness (Brandes)
- Closeness
- Harmonic closeness
- Eigenvector
- HITS
- CELF influence maximization
- Triangle count
- K-core decomposition
- Louvain
- Leiden
- Label propagation
- Infomap
- Walktrap
- Girvan-Newman
- Weakly connected components
- Strongly connected components
- Dijkstra
- A*
- Bellman-Ford
- Yen's k-shortest paths
- Max flow / min cut
- Steiner tree
- Minimum spanning tree
- Topological sort
- Node2Vec
- GraphSAGE
- Fast random projection
- Node similarity (Jaccard)
PostgreSQL + Apache AGE
No graph-algorithm library; AGE 1.8.0 adds C set-returning functions for unweighted (hop-count) BFS shortest paths callable from SQL or inside cypher().2518
Delivered as: extension
- shortest_path (unweighted BFS)
- all_shortest_paths (unweighted BFS)
- create_subgraph (induced-subgraph materialization)
Where can each one run?
GraphForge runs in-process only. PostgreSQL + Apache AGE: embedded not established, server yes, cluster not established.
| Mode | GraphForge | PostgreSQL + Apache AGE |
|---|---|---|
| Embedded / in-process | Shipped In-process library; no server, daemon, or connection string. 9 | Not established Runs inside a PostgreSQL server process; no embedded mode documented. |
| Server mode | Planned Optional remote peer extension planned after 0.6.0. 11 | Yes Installed into a PostgreSQL server with CREATE EXTENSION; Docker image provided. 17 |
| WebAssembly | Not planned A browser WebAssembly engine was declined in ADR 0027. 12 | Not established Not documented. |
| Distributed cluster | Not planned Distributed clusters and HA replication are out of scope. 9 | Not established No sharded/distributed graph deployment documented by AGE. |
| Managed cloud | Not planned No hosted compute. The graphforge.sh Hub distributes repositories; execution stays local. 13 | Partial No AGE-run cloud; offered as an extension on Azure Database for PostgreSQL flexible server. 22 |
| Replication | Not planned HA replication is out of scope. 9 | Yes Inherited from PostgreSQL physical/streaming replication; not AGE-specific. 26 |
What will my team write?
GraphForge: openCypher from Python, Node.js, Rust. PostgreSQL + Apache AGE: openCypher (via cypher() SQL function), SQL.
AGE developers embed Cypher strings inside SQL, with subquery and column-definition rules, from Go, Java, Node.js, or Python drivers. GraphForge developers write openCypher directly and call verbs that return Arrow.
- Query languages
- GraphForge openCypher
- PostgreSQL + Apache AGE openCypher (via cypher() SQL function), SQL
- SDKs and bindings
- GraphForge Python, Node.js, Rust
- PostgreSQL + Apache AGE Go, Java (JDBC), Node.js, Python
- Written in
- GraphForge Rust
- PostgreSQL + Apache AGE C (PostgreSQL extension)
- Surfaces
- GraphForge Python, Node.js, Rust crates, CLI, VS Code extension, Notebooks, Agent skills
What are the trade-offs?
AGE's analytics are unweighted shortest paths only. GraphForge has no SQL, joins, or multi-user server.
A fair assessment cuts both ways.
GraphForge
- No SQL interface and no joins with relational tables.
- Not a system of record for multi-user transactional applications.
- No server mode, managed hosting, or replication.
PostgreSQL + Apache AGE
- cypher() cannot be used as part of an SQL expression or as a standalone SELECT-list column; a subquery or FROM-clause call is required.21
- Built-in graph analytics are limited to unweighted shortest-path functions; multiple relationship types and non-zero min_hops raise feature-not-supported.25
- Docs recommend no DML or DDL be run directly in a graph's reserved namespace, so graph tables should be modified through Cypher/AGE functions.19
How would I migrate?
Load PostgreSQL + Apache AGE data through Arrow, pandas, or Polars. Standard openCypher carries over; procedure calls become verb calls.
Into GraphForge
- GraphForge 0.5.x loads data through Arrow, pandas, and Polars bulk calls or Cypher scripts; it has no native CSV loader.15
- Algorithms are API verbs (rank, cluster, paths, analyze, similar), not Cypher CALL procedures, so procedure calls become verb calls.7
- GraphForge passes the full openCypher 2024.3 TCK, so standard openCypher queries are the portable part of a migration.10
Moving from PostgreSQL + Apache AGE
- Vertices and edges can be bulk-loaded from CSV files with load_labels_from_file and load_edges_from_file.28
- Queries written in openCypher port from other Cypher databases, but must be wrapped in SELECT * FROM cypher(...) AS (...) with explicit agtype result columns.21
- Because graphs are ordinary PostgreSQL tables, standard PostgreSQL backup, replication and tooling apply to graph data.19
What else do people ask about PostgreSQL + Apache AGE?
Does Apache AGE include graph algorithms? AGE has BFS shortest paths only; its Cypher carries over; a dedicated engine fits when algorithms matter.
Does Apache AGE include graph algorithms?
AGE 1.8.0 added unweighted BFS shortest_path and all_shortest_paths functions plus create_subgraph(). There is no centrality or community-detection library.
Can I use AGE's Cypher queries in GraphForge?
The Cypher inside cypher() calls usually carries over, since both target openCypher. The SQL around it, such as column definition lists and joins with relational tables, does not.
Should I use PostgreSQL + AGE or a dedicated graph engine?
Use AGE when operational consolidation matters most and the graph queries are pattern matching or short paths. Use a dedicated engine when algorithms and analytical performance are the point.
How was this researched?
From 28 primary sources, accessed 2026-10-19, each cited where it is used.
Capabilities were checked against primary documentation, repositories, and release notes, accessed 2026-10-19. Each cell records two things: the capability’s state and the evidence behind it. Every GraphForge cell carries a release status. Shipped means stable 0.5.2; experimental means 0.6.0 release candidates only. GraphForge is pre-1.0 and makes no backward-compatibility guarantee yet.
- Yes documented natively
- Partial limited or caveated
- Extension official extension or module
- Paid edition paid or enterprise edition only
- Removed removed, with the version cited
- Not established not confirmed by our research; not a “no”
- Shipped in stable 0.5.2
- Partial shipped with documented limits
- Experimental in 0.6.0 release candidates only
- Planned on the roadmap
- Not offered outside the product's model
- Not planned explicitly out of scope
Benchmark classes
- Equivalent workloads. The same logical operation with the same correctness requirement on every engine.
- Native optimized workloads. Each engine's best supported implementation, including projections, indexes, extensions, and native algorithms.
- End-to-end application workloads. Time from raw data to a usable result: import, indexing, projection, computation, and output retrieval.
Sources · accessed 2026-10-19
- GraphForge repository README — GitHub
- GraphForge releases — GitHub
- GraphForge 0.5.0 changelog — GitHub
- Storage architecture — GraphForge Book
- Analytics integration: search and retrieval — GraphForge Guide
- ADR 0018: acknowledged durability and isolation — GitHub
- Algorithm verbs — GraphForge Book
- Cypher guide — GraphForge Guide
- openCypher compatibility — GraphForge Reference
- openCypher TCK compliance — GraphForge Reference
- Remote authority peer extension (#215) — GitHub
- ADR 0027: native runtime boundary — GitHub
- Release scope and roadmap — GraphForge Docs
- Keep research history — GraphForge Guide
- Graph construction — GraphForge Guide
- Knowledge public API v1 — GraphForge Book
- apache/age README — GitHub
- Release PG18/v1.8.0-rc0 (Apache AGE 1.8.0 release notes) — GitHub
- Graphs — Apache AGE manual
- MATCH — Apache AGE manual
- AGE Cypher query format — Apache AGE manual
- Apache AGE Extension — Azure Database for PostgreSQL, Microsoft Learn
- Full Text Search — PostgreSQL documentation
- pgvector — GitHub
- Add shortest_path / all_shortest_paths SRFs (PR #2430) — apache/age GitHub
- High Availability, Load Balancing, and Replication — PostgreSQL documentation
- The Apache Software Foundation Announces Apache AGE as a Top-Level Project — ASF blog
- Importing graphs from files — Apache AGE manual
Try GraphForge on your own graph.
Install the package, open a project, and run the same notebook we measure with.