Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is just a small experiment to start enabling RAG workloads with Eole.
It should serve as a basis for future iterations, such as enabling the management and use of such RAG workflows through the server for instance.
E.g. we could manage ChromaDB collections at the same level as models, and allow for RAG-specific prompts on top of the standard chat templates.
The example setup is using the rather hard to digest "EU AI Act" full text for the sake of the exercise.
Example
Note: This is not tuned at all. For now, we just grab the first result from the vector search, retrieve its context (previous and following chunks), and infer with some arbitrary settings.
Query:
Generated answer based on queried context: