Chromadb

Loader

Chromadb is an open-source embedding (vector) database. It lets you store documents alongside their embeddings and metadata, and query them by similarity, which makes it a common destination for retrieval-augmented generation (RAG) and other AI workflows. Connect it as a destination to load your data into a Chromadb collection.

Note: Chromadb is a loader, not a data source: it doesn't have its own streams, sync type or custom queries to sync from. The facts below reflect that.

At a glance

Property	Value
Type	Loader (target)
Authentication	None listed

What it loads

Whatever streams your connected data sources send it, written as documents into the Chromadb collection you configure.

Prerequisites

No credentials are listed for this connector. The following settings are validated together:

• Collection Name: the name of the collection to connect to in the Chromadb API (defaults to vector-db).
• Database Directory: the directory where the database files are stored (defaults to output).
• Document Text Property: the name of the property in the document that contains the text to be indexed (defaults to page_content).

Setup

In Chromadb

1. No credentials or account details are listed for this connector.

In Meltano Cloud

1. Enter the Collection Name (default: vector-db).
2. Enter the Database Directory (default: output).
3. Enter the Document Text Property (default: page_content).
4. Optionally set the Embeddings Property (default: embeddings) and the Metadata Property (default: metadata).

Available streams

This target doesn't discover streams itself: it loads whatever streams the connected extractor(s) send it, writing each record as a document in the configured collection.

Advanced configuration

• Flattening: Flattening Enabled sets whether nested objects in the document are flattened; Flattening Max Depth sets the maximum depth to which nested objects are flattened.
• Stream Maps / Stream Map Config: the list of stream maps to be used for indexing and querying, and the configuration for the stream map.

Settings

Setting	Type	Description
collection_name	string	The name of the collection to connect to in the Chromadb API. Default: vector-db.
db_directory	string	The directory where the database files are stored. Default: output.
document_text_property	string	The name of the property in the document that contains the text to be indexed. Default: page_content.
embeddings_property	string	The name of the property in the document that contains the precomputed embeddings. Default: embeddings.
metadata_property	string	The name of the property in the document that contains metadata. Default: metadata.
flattening_enabled	boolean	Whether or not to flatten nested objects in the document.
flattening_max_depth	integer	The maximum depth to which nested objects should be flattened.
stream_maps	object	The list of stream maps to be used for indexing and querying.
stream_map_config	object	The configuration for the stream map.

Need help?

If a stream or field you need isn't listed here or the connector doesn't work as expected, file it through the usual Meltano support channel.