Vectara
Use this guide to set up the Vectara destination connector and find reference information.
Vectara is a trusted GenAI platform providing Retrieval Augmented Generation, or RAG, as a service.
Connect any Airbyte source to Vectara and ingest data into your Vectara RAG pipeline.
:::info If you need support, use these public channels:
- For Airbyte-related issues, such as data sources or processing: Open a Github issue
- For Vectara-related issues, such as data indexing or RAG: Create a post in the Vectara forum or contact Vectara’s Discord server
:::
Overview
The Vectara destination connector supports Full Refresh Overwrite, Full Refresh Append and Incremental Append.
Output schema
All streams are output to a Vectara corpus. You must specify the corpus name in the config.
Vectara corpus names have no restrictions. If the specified corpus isn’t found, the connector creates a new corpus with that name. If multiple corpora share that name, Airbyte returns an error because it can’t determine the preferred corpus.
Supported sync modes
| Sync mode | Supported? |
|---|---|
| Full Refresh - Overwrite | Yes |
| Full Refresh - Append | Yes |
| Full Refresh - Overwrite + Deduped | No |
| Incremental Sync - Append | Yes |
| Incremental Sync - Append + Deduped | No |
Get started
You need a Vectara account to use Vectara with Airbyte. Follow these steps:
- Sign up for a Vectara account if you don’t already have one. After signing up, you’ll receive a Vectara customer ID. Find it by selecting your name in the top-right of the Vectara console window.
- Create a corpus within your account. A corpus is an area that stores text data you want to ingest into Vectara.
- Select the “Create Corpus” button in the console to create a corpus. Then provide a corpus name and description. Select your created corpus to see its name and corpus key at the top. See guide for more details.
- You can optionally define filtering attributes and apply advanced options.
- For the Vectara connector to work correctly, you must define a special string-typed meta-data field called
_ab_stream. The connector uses this field to identify source streams.
- The Vectara destination connector uses OAuth2.0 Credentials, an app client created in the Vectara console under API access > App clients. Keep your
Client IDandClient Secretready for connector setup.
Set up Vectara in Airbyte
You now have everything required to configure Vectara as a destination in the UI.
You’ll need this information to configure the Vectara destination:
- (Required) OAuth2.0 Credentials
- (Required) Client ID
- (Required) Client Secret
- (Optional) Customer ID. This is retained for backwards compatibility and is no longer needed to authenticate against the Vectara API v2.
- (Required) Corpus Name. Specify a corpus name you’ve set up manually using the instructions above. Alternatively, specify a corpus name that doesn’t exist. The connector creates a new corpus with that name and sets up the required meta-data filtering fields within it.
In the connector UI, you also define two sets of fields:
- Use
text_fieldsto define source fields that become text in Vectara for queries or summarisation. - Use
title_fieldto define the source field used as the document title in Vectara. - Use
metadata_fieldsto define source fields added to each document as meta-data.
Namespace support
This destination supports namespaces.
Changelog
Expand to review
| Version | Date | Pull Request | Subject |
|---|---|---|---|
| 0.3.0 | 2026-11-03 | PR | Upgrade to Python 3.13 (python-connector-base 4.1.1) and airbyte-cdk 7 |
| 0.2.32 | 2026-10-07 | PR | Migrate from the retired Vectara API v1 (and per-customer Cognito OAuth endpoint) to API v2 / auth.vectara.com |
| 0.2.31 | 2024-11-25 | 48659 | Update dependencies |
| 0.2.30 | 2024-11-04 | 48222 | Update dependencies |
| 0.2.29 | 2024-10-29 | 47744 | Update dependencies |
| 0.2.28 | 2024-10-23 | 47084 | Update dependencies |
| 0.2.27 | 2024-10-12 | 46812 | Update dependencies |
| 0.2.26 | 2024-10-05 | 46438 | Update dependencies |
| 0.2.25 | 2024-09-28 | 46114 | Update dependencies |
| 0.2.24 | 2024-09-21 | 45806 | Update dependencies |
| 0.2.23 | 2024-09-14 | 45481 | Update dependencies |
| 0.2.22 | 2024-09-07 | 45324 | Update dependencies |
| 0.2.21 | 2024-08-31 | 45021 | Update dependencies |
| 0.2.20 | 2024-08-24 | 44657 | Update dependencies |
| 0.2.19 | 2024-08-22 | 44530 | Update test dependencies |
| 0.2.18 | 2024-08-17 | 44310 | Update dependencies |
| 0.2.17 | 2024-08-12 | 43858 | Update dependencies |
| 0.2.16 | 2024-08-10 | 43494 | Update dependencies |
| 0.2.15 | 2024-08-03 | 43153 | Update dependencies |
| 0.2.14 | 2024-07-27 | 42705 | Update dependencies |
| 0.2.13 | 2024-07-20 | 42168 | Update dependencies |
| 0.2.12 | 2024-07-13 | 41829 | Update dependencies |
| 0.2.11 | 2024-07-10 | 41362 | Update dependencies |
| 0.2.10 | 2024-07-09 | 41140 | Update dependencies |
| 0.2.9 | 2024-07-06 | 40953 | Update dependencies |
| 0.2.8 | 2024-06-27 | 40215 | Replaced deprecated AirbyteLogger with logging.Logger |
| 0.2.7 | 2024-06-25 | 40321 | Update dependencies |
| 0.2.6 | 2024-06-22 | 39973 | Update dependencies |
| 0.2.5 | 2024-06-06 | 39193 | [autopull] Upgrade base image to v1.2.2 |
| 0.2.4 | 2024-05-20 | 38432 | [autopull] base image + poetry + up_to_date |
| 0.2.3 | 2024-03-22 | #37333 | Updated CDK & pytest version to fix security vulnerabilities |
| 0.2.2 | 2024-03-22 | #36261 | Move project to Poetry |
| 0.2.1 | 2024-03-05 | #35206 | Fix: improved title parsing |
| 0.2.0 | 2024-01-29 | #34579 | Add document title file configuration |
| 0.1.0 | 2023-11-10 | #31958 | 🎉 New Destination: Vectara (Vector Database) |