YAGO 3.10 Knowledge Graph

Large heterogeneous knowledge graph automatically extracted from Wikipedia and Wikidata. Contains 5 node types: entities, types, literals, relations, and concepts. Contains 37+ relation types (isA, created, hasCapital, etc.). Serves as a reference knowledge base for information retrieval and entity linking. Demonstrates multi-relational graph structure typical of Linked Open Data.

npx keepset fetch yago3-10

Nodes: 123,182
Edges: 1,089,040
Size: 18.4 MiB

Fetch and preview this dataset on this page, or download the result as JSON. The command line fetches it without a browser.

Overview

Id	`yago3-10`
Category	`graph/v1`
Kind	graph
Domain	Knowledge graph
Licence	not recorded by the source
Catalogue	`keepset` 1.33.0

Sources and provenance

#	URL	Status	Retrieved	Read by recipe	Pinned snapshot	SHA-256
0	https://web.informatik.uni-mannheim.de/pi1/kge-datasets/yago3-10.tar.gz	live	2026-03-26	yes
1	https://www.yago-knowledge.org/	live	2026-03-26

Citations

Suchanek et al. (2007). YAGO: A Core of Semantic Knowledge Unifying DBpedia and Wikipedia. ACM International Conference on World Wide Web.

Recipe

Source	#0
Archive	tar.gz
Part train	`yago3-10/train.txt`
Part valid	`yago3-10/valid.txt`
Part test	`yago3-10/test.txt`
Parser format	`"kg-triples"`
Parser encoding	`"utf-8"`
Parser delimiter	`"tab"`
Parser order	`"head-relation-tail"`
Parser header	`"none"`
Parser quoted	`false`

Crosswalk

keepset/relations-split to graph/v1, as a crossmap crosswalk.

Rule	From	To	Direction
move	`meta`	`meta`	both
move	`nodes[].id`	`nodes[].id`	both
move	`edges[].source`	`edges[].source`	both
move	`edges[].target`	`edges[].target`	both
move	`edges[].type`	`edges[].type`	both
move	`edges[].split`	`edges[].split`	both

Expected counts

Edges	1,089,040
Nodes	123,182

Properties

Components	11
Connected	no
Cyclic	yes
Density	0.0004194
Directed	yes
Download bytes	18.4 MiB
Edge types	37
Heterogeneous	yes
Multigraph	yes
Node types	1
Raw bytes	0 B
Self loops	no
Weighted	no