embedded Datalog / multi-model database
Datalevin is an embedded, ACID, multi-model durable database that exposes a Datalog query engine over key-value, document, graph, full-text and vector data, shipped as a Clojure/JVM library.
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
Choose when you need MySQL-wire-compatible distributed SQL with HTAP scale-out instead of an in-process library.
TiDB is built for agentic workloads that grow unpredictably, with ACID guarantees and native support for transactions, analytics, and vector search. No data silos. No noisy neighbors. No infrastructure ceiling.
Choose when you need geo-distributed Postgres-compatible SQL with automatic failover rather than an embeddable engine.
CockroachDB — the cloud native, distributed SQL database designed for high availability, effortless scale, and control over data placement.
Choose when you want managed serverless Postgres with branching and scale-to-zero instead of a database running in your process.
Neon: Serverless Postgres. We separated storage and compute to offer autoscaling, code-like database branching, and scale to zero.
Choose when you want SQLite semantics behind a Raft cluster for HA, accepting network hops and SQL over Datalog.
The lightweight, fault-tolerant database built on SQLite. Designed to keep your data highly available with minimal effort.
Choose when you need a mature Cypher graph server with tooling, at the cost of a JVM server process and GPL-3.0 licensing.
Graphs for Everyone
Choose when you are already on Postgres and need to shard it horizontally rather than adopt a new embedded engine.
Distributed PostgreSQL as an extension
Choose when you need a distributed C++ graph store at billion-edge scale, not a single-process embedded database.
A distributed, fast open-source graph database featuring horizontal scalability and high availability
Choose when you need distributed, Postgres-compatible SQL with strong consistency across regions.
YugabyteDB - the cloud native distributed SQL database for mission-critical applications.
Choose when you want a multi-model graph/document/full-text server with SQL-ish queries and ACID, accepting JVM server weight.
OrientDB is the most versatile DBMS supporting Graph, Document, Reactive, Full-Text and Geospatial models in one Multi-Model product. OrientDB can run distributed (Multi-Master), supports SQL, ACID Transactions, Full-Text indexing and Reactive Queries.
Choose when your graph data already lives in Postgres and you want Cypher queries layered on as an extension.
Graph database optimized for fast analysis and real-time data processing. It is provided as an extension to PostgreSQL.
Choose when you need built-in data versioning/branching and WOQL-style reasoning on a shared server.
TerminusDB is a distributed, collaborative database designed for building, sharing, versioning, and reasoning on structured data.
Choose when you want a lightweight embedded Rust graph datastore with pluggable backends but no Datalog rules.
A graph database written in rust
Choose when you want an embedded Go graph/KV store with its own simple query language and no JVM dependency.
EliasDB a graph-based database.
Choose when you need a distributed ACID document store on FoundationDB rather than an embedded one.
Distributed, transactional document database backed by FoundationDB.
Choose when you want an embedded MongoDB-style document store in Rust and do not need Datalog rules or graph queries.
PoloDB is an embedded document database.
Choose when you want an embedded Cypher property-graph engine with vector and full-text search, though it is now archived.
Embedded property graph database built for speed. Vector search and full-text search built in. Implements Cypher.
Choose when you want an embedded NoSQL document store on plain JVM/Android without Datalog recursion.
NoSQL embedded document store for Java
Choose when you need an embeddable multi-modal store with Python/Pandas/NetworkX bindings instead of a Clojure Datalog one.
Multi-Modal Database replacing MongoDB, Neo4J, and Elastic with 1 faster ACID solution, with NetworkX and Pandas interfaces, and bindings for C 99, C++ 17, Python 3, Java, GoLang 🗄️
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
Closest peer (Datalog over relational-graph-vector storage) but unmaintained for ~2 years, so pick it only if you can fork and patch it yourself.
A transactional, relational-graph-vector database that uses Datalog for query. The hippocampus for AI!
An in-memory Datalog interpreter for TypeScript with no durable storage; usable only for query evaluation.
a datalog interpreter in typescript
These projects were analysed and named datalevin among their alternatives. The relationship is not symmetric — how datalevin rates them is a separate judgement, made when datalevin is analysed in its own right.
calls datalevin “embedded Datalog database; choose for a simple local knowledge base with Datalog queries.”
🥑 ArangoDB is a native multi-model database with flexible data models for documents, graphs, and key-values. Build high performance applications using a convenient SQL-like query language or JavaScript extensions.