Skip to content

Indexed Collection (for RAG applications)

When should I use this?

When you want your bot's responses to be grounded in your uploaded documents.

Definition

Retrieval-Augmented Generation (RAG) is a technique where a language model retrieves relevant information from a set of documents to ground its answers in real data. Instead of relying solely on its built-in knowledge, the model uses indexes—specialized databases that store document content as vectors (numerical representations of meaning). This makes it easy for the model to find and use the most relevant parts of your uploaded files when answering questions.

Indexed collections will replace OpenAI Assistants' file search functionality in the future

Consult the migration guide if you have assistants that you want to replace with indexed collections.

If you’ve used the OpenAI Assistants’ file search capability in OCS, you’ve already interacted with an index behind the scenes.

In OCS, there are two types of indexes:

  • Remote Index
  • Local Index

Remote Index

Remote indexes are hosted and managed by an LLM provider. Files and index configuration are uploaded to the provider, which maintains and manages the index. The embedding model used to create file embeddings is selected by the provider.

Supported providers

OpenAI Remote Collection Limit

When using OpenAI as your LLM provider with remote (OpenAI-hosted) indexed collections, you can select a maximum of 2 collections per LLM node. This is a limitation imposed by OpenAI's API, not Open Chat Studio.

  • Local indexes are NOT affected by this limit
  • Non-OpenAI providers are NOT affected by this limit
  • If you attempt to select more than 2 remote collections with OpenAI, you will receive a validation error

If you need to use more than 2 collections, consider using local indexes instead.

Supported file types

Supported files are determined by the selected provider:

Local Index

Local indexes are a new feature in OCS. We are actively working to support additional file types and embedding models, allowing you to better customize your index with models that suit your needs.

Local indexes are hosted and managed by OCS. When you create a local index, you can select which embedding model to use for generating file embeddings. Embedding models are provided by LLM providers. Just as different LLM models have varying strengths and weaknesses, different embedding models also have their own strengths and weaknesses. Thus, choosing the right embedding model is important to ensure the best performance for your specific use case.

Supported providers

  • OpenAI

Supported file types

  • pdf
  • txt
  • csv
  • docx

Supported embedding models

You can see the supported embedding models for each provider when creating or editing the provider in your team settings.

Chunking and Optimization

When you upload a document to an index, it’s broken up into smaller parts called chunks. These chunks are then converted into vectors and stored in the index. Chunking is a key part of how RAG works, as it affects how accurately the model can retrieve relevant information.

In most cases, it will not be necessary to change the default chunking strategy, but you’ll have the option to customize the chunking strategy for each set of uploaded files:

  • Chunk size – how large each chunk is (in tokens)
  • Chunk overlap – how much each chunk overlaps with the next, to preserve context

Choosing the right chunking strategy can improve retrieval accuracy, especially for technical documents.

This flexibility helps tailor the index to your use case—whether it’s short notes or long, complex reports.

Document Sources

In addition to manually uploading documents to a collection, you can also configure document sources from which Open Chat Studio will automatically load and index documents.

The primary advantage of document sources over manual uploads is that Open Chat Studio can check for updates periodically, which eliminates the need for manual updates.

The following document source types are currently supported:

Confluence

Load pages from a Confluence site. Pages can be filtered using the space key, label, CQL, or individual page IDs.

Authentication

Use a Basic Auth authentication provider with your Atlassian username and use your API Key as the password.

Configuration

Field Description
Site URL The URL of the Confluence site (e.g. https://yoursite.atlassian.net/wiki)
Max Pages The maximum number of pages to load
Space Key Load pages from this space
Label Load pages with this label
CQL CQL query to use to search for pages to load
Page IDs Load only these specific pages

Note

Only one of the Space Key, Lable, CQL and Page IDs fields can be used at a time.

GitHub

Load pages from a GitHub repository. Files can be filtered by path and by matching patterns against the filenames.

Authentication

Use a Bearer Token authentication provider.

Configuration

Field Description
Repository URL GitHub repository URL (e.g. https://github.com/user/repo)
Branch Git branch to sync from
File Pattern File patterns to include. Prefix with '!' to exclude matching files.
Path Filter Optional path prefix to filter files (e.g., docs/)

Monitoring Sync Status

Open Chat Studio tracks the history of every sync run for each document source. Use the sync logs to confirm that syncs are completing successfully and to diagnose problems when they are not.

Status indicator

Each document source header displays a status indicator reflecting the outcome of the most recent sync:

  • Error – the last sync encountered a problem. The indicator is shown in red.
  • Success – the last sync completed without errors.
  • In progress – a sync is currently running. The indicator animates to show activity.