Azure

At a glance

The community member is having issues integrating RAG in llama-index with Azure cloud, specifically Azure Cosmos DB NoSQL or Azure Cosmos DB for Mongo. They are using the example from the docs for Multi-Document Agents (V1), with retrieval on Nodes and mapping to Documents. The community member encountered two problems: A) they got stuck when uploading Nodes, whereas Documents work; B) the llama-index-vector-stores-azurecosmosnosql package seems to be old and has dependency issues.

In the comments, another community member suggests providing a code example to help with the nodes vs documents issue. They also mention that the vector store package doesn't seem that old, as it only has one main dependency, and they welcome a pull request to update it if it's out of date.

Useful resources

aalessandro

Hello, this question is about integrating RAG in llama-index with Azure cloud, in particular the cloud DB (Azure Cosmos DB NoSQL or Azure Cosmos DB for Mongo).
My RAG is based on the example in the docs Multi-Document Agents (V1 ), with retrieval on Nodes and mapping to Documents. I gave Azure Cosmos DB NoSQL a try but A) I got stuck when uploading Nodes, whereas Documents work; B) llama-index-vector-stores-azurecosmosnosql seems to be old and get into dependency issues. Am I missing something?

2 comments

LLogan M

You might need to give a code example for the nodes vs documents thing

LLogan M

The vector store isn't that old? It only has one main dependency
https://github.com/run-llama/llama_index/blob/d4a31cf6ddb5dd7e2898ad3b33ff880aaa86de11/llama-index-integrations/vector_stores/llama-index-vector-stores-azureaisearch/pyproject.toml#L33

If that's out of date, I welcome a PR to bump it (i don't have access to test it)

Add a reply

Find answers from the community

Azure