RAG (Retrieval-Augmented Generation)

With Acris, you can build a RAG (Retrieval-Augmented Generation) app to answer questions on your data.

  1. We first upload our files to Acris.
  2. Then we create a step and select the Semantic Search tool to retrieve the relevant information.

Upload Files

We go to the Sources tab and upload our files.

RAG Chatbot

Search on Files

We then create a new step and add instructions to it.

The content of the step will be sent to a Vector Database to do a semantic search.

RAG Chatbot

Run

When we run the step, it will ask us to fill in the query. Here we can ask questions on the content of the files.

RAG Chatbot

Result

The result will be a list of the most relevant information and the answer to the question.

RAG Chatbot

Sources

We can also see the sources of the information that was used to answer the question.

RAG Chatbot

Now, let's check another use case here