Is attaching data source at Chat completion Azure Ai Foundary permenant

Mohamed Hussein 710 Reputation points
2025-03-07T04:06:18.1933333+00:00

Good Day

Is attaching data source at Chat completion Azure Ai Foundary permenant?

By other words, Can i attach files from the playground, then call Chat completions APIs directly?

or it's related to active session only?

Is the created vector index permenant? or i should attache it with every single call?

User's image

Azure OpenAI Service
Azure OpenAI Service
An Azure service that provides access to OpenAI’s GPT-3 models with enterprise capabilities.
4,108 questions
{count} votes

1 answer

Sort by: Most helpful
  1. SriLakshmi C 6,250 Reputation points Microsoft External Staff Moderator
    2025-03-07T07:16:45.7433333+00:00

    Hello Mohamed Hussein,

    Is attaching data source at Chat completion Azure Ai Foundry permanent other words, can I attach files from the playground, then call Chat completions APIs directly? or it's related to active session only?

    Attaching a data source in Azure AI Foundry is not permanent and is linked to the active session only. When you attach files from the playground, the connection remains active for the duration of that session but does not persist beyond it. This means that for each new session or API call, you must reattach the data source. If you need to use the same data source in multiple API calls, you will have to explicitly reference or reattach it each time.

    Is the created vector index permanent? or I should attach it with every single call?

    In Azure AI Foundry, both data source attachments and vector indexes are not permanent and are tied to the active session. When you attach files from the playground, the connection remains active only for that session, requiring you to reattach the data source for each new session or API call. Similarly, the vector index you create is not automatically permanent and must be referenced in every call to ensure the model can access the relevant data. However, if you are using a persistent vector index, such as Azure AI Search or another vector database, the index remains stored, and you can save the configuration for reuse. Still, you must explicitly reference it in each API call, especially when performing retrieval-augmented generation (RAG).

    I hope you understand. And, if you have any further query do let us know.

    Thank you!

    1 person found this answer helpful.
    0 comments No comments

Your answer

Answers can be marked as Accepted Answers by the question author, which helps users to know the answer solved the author's problem.