Skip to main content
Version: main

Supported Data Sources

TypeCRUDCatalogNotesDocs
PostgreSQLFullYespgvector KNN, FTSdocs
MySQLFullYesTable or catalog registrationdocs
SQLiteFullYessqlite-vec KNN, FTSdocs
MongoDBFullNoCollections with point lookupsdocs
RedisFullNoHashes mapped to SQL rowsdocs
DynamoDBFullYesScan + filter pushdowndocs
SeekDBFullYesMySQL-wire CRUD, FULLTEXT, HNSWdocs
ClickHouseReadYesColumnar OLAP, filter/limit pushdowndocs
LanceRead + job-writeNoKNN, BM25 FTS; job destinationdocs
Apache IcebergReadNoSchema evolution, partition pruningdocs
InfluxDB 3ReadNoTime series over Arrow Flight SQLdocs
S3 / GCS / AzureReadNoCSV, Parquet, Lance in object storesdocs
CSV / ParquetReadNoLocal or remote filesdocs
SaaS via Open ConnectorReadYesGitHub, Slack, Notion, Feishu, Gmail, Discord, Outlook, OneDrive, Google Drive, Dropbox packs as stable SQL tables; pushdown + TTL cachedocs
DocumentsReadNoPDF/Office/ODF/image → per-page Markdown, tables, imagesdocs
RSS / AtomReadYesFeeds as feeds + items; per-feed TTL cache, fault isolation, un-sandboxed fetch egressdocs
Graph (Apache AGE)ReadYesRead-only openCypher over Postgres; YAML views as catalog tables, cypher_query UDTFdocs