defenseunicorns / leapfrogai

Production-ready Generative AI for local, cloud native, airgap, and edge deployments.
https://leapfrog.ai
Apache License 2.0
244 stars 25 forks source link

feat(api) Support Uploading Word Files for RAG #620

Open barronstone opened 2 weeks ago

barronstone commented 2 weeks ago

User Story

As a LeapfrogAI end user I want to upload Microsoft Word (.doc and .docx) files for RAG So that I can create an Assistant to "chat with my doc"

Acceptance Criteria

Given a Microsoft Word file When the user uploads it to LeapfrogAI via the UI or the API Then the file will be chunked and vectorized for RAG

Additional context

The initial version of this feature may ignore embedded images within the Word document. Another lower-priority issue will be opened to investigate using OCR or some other form of image analysis to extract information from embedded images.