adithya-s-k/VARAG
A vision-first RAG engine that uses vision-language models for multimodal document retrieval and generation.
Not currently ranked — collecting fresh signals.
star history
VARAG implements retrieval-augmented generation with a visual focus, integrating vision-language models to process and index both visual and textual content from documents. It supports techniques including ColPali-based retrieval, OCR through Docling, and vector-based semantic search to enable grounded responses from multimodal sources.
Frequently asked
- What is adithya-s-k/VARAG?
- A vision-first RAG engine that uses vision-language models for multimodal document retrieval and generation.
- Is VARAG open source?
- Yes — adithya-s-k/VARAG is an open-source project tracked on heatdrop.
- What language is VARAG written in?
- adithya-s-k/VARAG is primarily written in Python.
- How popular is VARAG?
- adithya-s-k/VARAG has 497 stars on GitHub.
- Where can I find VARAG?
- adithya-s-k/VARAG is on GitHub at https://github.com/adithya-s-k/VARAG.