AIDB 3.0.1 release notes v7
Released: 03 Apr 2025
AIDB 3.0.1 includes the following enhancements and bug fixes:
| Type | Description |
|---|---|
| Enhancement | Live auto processing for the preparer pipeline: The preparer pipeline now supports live auto processing for data sources that are Postgres tables. When enabled, AIDB creates triggers on the source table to automatically run the preparer task. |
| Enhancement | Model encode/decode batch functions now return rows: The functions aidb.decode_text_batch() and aidb.encode_text_batch() now return a set of rows instead of a single row containing an array. This change was made based on customer feedback. Batch functions now return one row per input value. |
| Enhancement | Auto-processing now configured via new functions: The functions aidb.enable_auto_embedding_for_table() and aidb.disable_auto_embedding_for_table() are deprecated. The new function aidb.set_retriever_auto_processing() accepts a processing mode argument: Live and Disabled modes turn auto-embedding on and off. For the preparer pipeline, aidb.set_preparer_auto_processing() serves the same purpose. |
| Enhancement | Added support for pgfs.allowed_local_paths GUC: A new GUC pgfs.allowed_local_paths controls local file access in PGFS. By default, local file access is now restricted to /tmp/pgfs. This GUC can be used to allow access to other directories. |
| Enhancement | Retriever pipeline now automatically creates vector indexes: Vector indexes are now automatically created when creating retrievers. The aidb.create_retriever_for_table() and aidb.create_retriever_for_volume() functions now accept an optional index_type argument to specify the index type. The default value is vector. |