Details
- Vendor
- Alibaba
- Type
- Property Graph
- Category
- Growing
- License
- Apache-2.0
- Language
- C++
- Query Languages
- GremlinopenCypher
- Protocols
- BoltHTTP/RESTGremlin (WebSocket)gRPC
The go-to resource for graph technology details
A distributed graph computing platform from Alibaba combining three engines: GRAPE for analytics, an interactive engine for traversals, and Graph-Learn for graph neural networks. Exposes a unified Python interface that runs across a cluster of machines and supports both Gremlin and Cypher.
Protocols gRPC, HTTP/REST, Bolt, Gremlin (WebSocket)
Python client talks to the coordinator over gRPC for the analytical/learning engines. The Interactive (GIE/Flex) engine additionally exposes an HTTP/REST admin+query API (see graphscope.io/docs/flex/interactive/development/restful_api), a Bolt endpoint for Cypher via standard neo4j drivers, and a Gremlin WebSocket endpoint (ws://.../gremlin) for the GIE Gremlin frontend. Draft only listed gRPC and HTTP/REST.
you could deploy and manage the workloads of GraphScope through a python client, which communicates with the GraphScope engines on the K8s cluster through a gRPC serviceDeploy on K8s Cluster - GraphScope documentation · reviewed 2026-08-02
You can establish a connection to the Interactive service using Neo4j’s Bolt connector and execute Cypher queries.Getting Started - GraphScope documentation · reviewed 2026-08-02
AdminService API (with tag AdminService) defines the API for GraphManagement, ProcedureManagement and Service Management.GraphScope Interactive API v0.3 · reviewed 2026-08-02
Released 2020-12
Corrected from the catalogue's '2021' (likely conflated with the 2021 VLDB papers describing GraphScope): the source code was first publicly released on December 7, 2020, per GraphScope's own launch post, matching the GitHub repository's creation date of 2020-10-28 as a lead-up.
The source code of GraphScope is released today!Say Hello to GraphScope! · reviewed 2026-08-02
Language C++
GraphScope's own docs describe its core analytics execution engine (GAE, built on GRAPE) as written in C++, with Java and Python exposed via FFI; the codebase is genuinely split across C++, Rust and Java (GitHub byte counts are close between the three), so this reflects the engine the project leads with rather than a single unambiguous answer. Matches the current catalogue value.
The execution engine of GAE is written in C++ and supports other programming languages (i.e., Java and Python) through FFIs (Foreign Function Interfaces), allowing for better integration with other big data ecosystems.Overview and Architecture - GraphScope documentation · reviewed 2026-08-02
License Apache-2.0
SPDX identifier as classified by GitHub from the repository licence file.
"spdx_id": "Apache-2.0"GitHub licence API — alibaba/GraphScope · reviewed 2026-08-02