arcadedb.com

What It Is

ArcadeDB is an open-source multi-model graph database developed by Arcade Data Ltd. and licensed under Apache 2.0. Its core positioning is not simply as a graph database, but as an engine that natively supports graphs, documents, key-value data, full-text search, vectors, and time-series data in one place. This can reduce the synchronization and operations overhead that comes with combining multiple systems such as Neo4j, MongoDB, Elasticsearch, vector databases, and time-series databases.

Core Capabilities

From a developer tooling perspective, ArcadeDB’s appeal lies in its wide range of query and deployment options. At the query layer, it supports SQL, OpenCypher 25, Gremlin, GraphQL, MongoDB Query Language, and the Java API, while also being compatible with protocols such as HTTP/JSON, Postgres wire protocol, Redis, and MongoDB. For deployment, it supports Docker, Kubernetes, On Premise, and can also be embedded in Java applications, making it suitable for gradual evolution from local experiments to production clusters. For transactions, it provides ACID and WAL, and uses Raft to maintain consistency across multiple servers.

Pricing and Open Source

The official website clearly emphasizes that all features are free forever, with no enterprise edition and no cloud-only features. It also promises that the Apache 2.0 license will remain unchanged long term. This is valuable for companies concerned about database projects changing licenses or introducing commercial restrictions later. Monetization mainly comes from enterprise support subscriptions, SLAs, priority fixes, support from core developers, custom development, consulting, and sponsorships, but specific support pricing is not disclosed.

Pros and Cons

Its advantages include consolidated multi-model capabilities, broad query language coverage, flexible deployment options, and no open-core feature wall. The official site also provides scenario descriptions for graph analytics benchmarks, migration from OrientDB/Neo4j, Customer 360, Graph RAG, AI/ML Feature Store, and more. The limitations are that performance data still needs to be re-tested by users against their own workloads; enterprise support pricing is not transparent; and the main materials do not provide clear information about official cloud hosting, Chinese documentation, a domestic community, or a large-scale third-party ecosystem.

Who It’s For and Access from China

ArcadeDB is suitable for backend teams, data platform teams, and AI application teams that need self-hosting, strong license certainty, and unified modeling across multiple data models. It is especially relevant for mixed scenarios such as customer relationship networks, real-time recommendations, identity resolution, Graph RAG, vector similarity, and time-series behavior analysis. The source material does not provide details on access from China, so domain connectivity, GitHub/Docker image pulls, and payment methods for enterprise support all need to be tested in practice. If stronger domestic access or ecosystem requirements are important, alternatives such as Neo4j, ArangoDB, MongoDB, Elasticsearch, Weaviate, Pinecone, and TimescaleDB may be worth comparing.