Wikipedia
Collection
Embeddings indexes and datasets for Wikipedia data.
•
4 items
•
Updated
•
1
This is a txtai embeddings index for the English edition of Wikipedia.
The slim version has the 100K most popular
Wikipedia pages ranked by page views. This embeddings index also has graph indexing enabled, which enables using this as a source for GraphRAG.
See the txtai-wikipedia model page for additional information on this datasource.