News and technical deep dives from the ArcadeDB team. Every release is also listed in the Changelog.
Fraud rings, synthetic identities, and money laundering are connection problems, not row problems. How to model and query all three in ArcadeDB with graph traversal, vector similarity, full-text matching, and...
ArcadeDB 26.10.2 closes 682 issues and pull requests across about 1,360 commits. A query admission gate and list/terminate for running queries arrive, point lookups get 3 to 7 times faster...
ArcadeDB runs ISO GQL syntax inside its native Cypher 25 engine. Here is what works today, what does not yet, and how GQL and Cypher mix in one query.
ArcadeDB 26.10.1 closes 1,615 issues and pull requests across about 3,100 commits. Raft high availability is hardened end to end, lost updates and silent data loss are closed, index, SQL...
PostgreSQL 19 was going to ship SQL/PGQ graph queries, then reverted them. Even when they land, they compile to joins. We ran the LDBC LSQB benchmark: ArcadeDB finished all nine...
Tom Sawyer Software built its Digital Engineering suite, Planning, Business Process, Model-Based Engineering, Data Streams, Perspectives, and Explorations, on ArcadeDB as the open-source graph analytics and graph database backbone.
A new Arrow client stack spent three weeks exercising ArcadeDB's PostgreSQL wire protocol and found four gaps. The first was fixed in four and a half hours, an external pull...
ArcadeDB is now a first-class integration in Google's MCP Toolbox for Databases, the open-source MCP server that connects AI agents to enterprise databases alongside PostgreSQL, BigQuery, Spanner, Snowflake, and Neo4j....
ArcadeDB now ships four native drivers: HTTP and gRPC clients for Python and for TypeScript/JavaScript, all generated from one OpenAPI spec and one Protobuf contract, published on PyPI and npm...
ArcadeDB 26.9.1 is the largest release we have ever shipped: 992 issues and pull requests closed, 1,500 commits. Full backups are 27.6x faster and no longer stall writers, a family...
ArcadeDB 26.8.1 is a major release: 381 issues and pull requests closed, 669 commits. Concurrent writes to unrelated records of the same page no longer conflict, vertex deletes no longer...
ArcadeDB 26.7.3 is a focused hotfix on top of 26.7.2: it closes three security advisories (two in the MCP server transport, one in the JavaScript/Java trigger authorization gate) and repairs...
ArcadeDB 26.7.2 ships full Bolt driver certification: 39 conformance scenarios run nightly against the Java, JavaScript, Python, .NET, and Go Neo4j drivers across 14 pinned versions, published as a live...
ArcadeDB 26.7.2 is a stability and correctness release with 130+ resolved issues: a deep engine audit across storage, WAL recovery, transactions, LSM indexes and concurrency, full Neo4j Bolt driver certification,...
ArcadeDB adds OpenTelemetry tracing, RED latency metrics with OTLP export, structured JSON logging with correlation IDs, and Kubernetes health probes. Every one of them is opt-in and backward compatible.
ArcadeDB 26.7.1 is a large stability, resilience and security release with 420+ commits and 238 resolved issues: Raft/HA hardening with per-database quarantine and divergence self-recovery, OpenTelemetry distributed tracing, native BM25...
ArcadeDB 26.6.1 brings end-to-end TLS/SSL for HA clusters, a deep durability and crash-recovery hardening pass across the WAL and storage layers, a broad security hardening sweep, and a long list...
Step-by-step guide to deploying a high-availability ArcadeDB cluster on Kubernetes using the official Helm chart. Includes a kind-based local example, production values, Raft consensus, persistence, and verification.
ArcadeDB 26.5.1 ships a brand-new sparse vector index with server-side hybrid retrieval, INT8 quantization end-to-end, EXTERNAL property storage, query partitioning, offline HA cluster bootstrap, and an extensive batch of OpenCypher...
ArcadeDB passed 34 of 34 in-house Jepsen tests on its Raft (Apache Ratis) HA stack: ACID, linearizability, and transaction isolation under partitions, crashes, pauses, and clock skew.
ArcadeDB Academy is live: 6 free, self-paced courses covering fundamentals, SQL, Cypher graph queries, Neo4j migration, OrientDB migration, and Vector Search with RAG. 135 lessons, quizzes, and a professional certification....
ArcadeDB's new GraphBatch delivers up to 8.39x faster graph ingestion than the standard API and 3x faster than the previous GraphImporter, reaching over 1.2 million edges per second at medium...
How Cognee's AI memory engine and ArcadeDB's multi-model database work together to give AI agents persistent, structured memory — with a single backend for graphs, documents, and vectors.
ArcadeDB v26.3.2 introduces the Graph OLAP Engine with Compressed Sparse Row encoding, delivering up to 462x speedups on analytical workloads while keeping full OLTP performance. Benchmarks show ArcadeDB is now...
ArcadeDB 26.3.2 is a performance-focused release with 100+ commits and 21 closed issues, featuring a Graph Analytical View (OLAP), high-performance bulk edge creation, HTTP batch endpoint, gRPC batch loading, GraphQL...
The best open-source Neo4j alternatives in 2026 compared: ArcadeDB, ArangoDB, KuzuDB, Memgraph, FalkorDB, and HugeGraph — covering licensing, performance, multi-model support, and AI readiness.
ArcadeDB v26.3.1 introduces a built-in MCP server, letting you connect Claude Desktop, ChatGPT, or any MCP-compatible AI client directly to your database for natural language queries, schema exploration, and report...
ArcadeDB 26.3.1 is a major release with 190+ commits and 52 closed issues, featuring a redesigned Studio with AI Assistant, MCP Server, geospatial indexing, native time-series support, materialized views, built-in...
ArcadeDB now has an official LangChain integration package. Build Graph RAG applications with GraphCypherQAChain, import graph documents, and query your knowledge graph using natural language — all backed by ArcadeDB's...
Why ArcadeDB is the perfect KuzuDB replacement — with LDBC benchmark proof that ArcadeDB is up to 9x faster than Kuzu on graph analytics, plus a complete step-by-step migration guide....
As databases like ArangoDB, MongoDB, Redis, Elasticsearch, and CockroachDB abandon open source licenses, ArcadeDB commits to Apache 2.0 forever. Here's why license stability matters for your project.
ArcadeDB 26.2.1 adds a Custom Package Builder - a shell script that creates distributions containing only the modules you actually use, reducing size from 150MB to just what you need....
ArcadeDB migrates its SQL parser from JavaCC to ANTLR, delivering 75.9% faster parsing across all query types while maintaining very high compatibility with existing queries.
ArcadeDB 26.2.1 introduces automatic database backups with scheduled execution, intelligent retention policies, and zero-downtime operation through Studio or JSON configuration.
ArcadeDB introduces a native OpenCypher engine replacing the legacy Cypher-to-Gremlin translation layer, delivering 25x faster graph queries with a fraction of the dependencies.
A comprehensive look back at ArcadeDB's 2025 milestones including JVector integration, Python bindings, gRPC support, RID recycling, and Java 21 migration.
ArcadeDB 25.3.1 brings complete space reuse and RID recycling, performance improvements, new database management commands, and PostgreSQL compatibility enhancements.
ArcadeDB v21.12.1 introduces the new GraphQL module, database check/repair command, and node alignment for distributed clusters.
ArcadeDB v21.11.1 brings real-time WebSocket event listeners, Gremlin 3.5.1 upgrade with embedded Gremlin Server, and new Redis persistent and transient commands.
ArcadeDB 21.10.1 brings faster Gremlin and Cypher queries with index optimization, enhanced server security with user groups, and SQL support for backup, import and export operations.
Introducing ArcadeDB - the next generation multi-model database. Learn about the vision, the Arcade Business Model, and why we built a brand new transactional engine from the ground up.