- Published on
'A deep dive into my recent research on spatially-grounded document retrieval using ColPali models and OCR bounding boxes, enabling precise region-level retrieval during inference time and without additional training.'
Latest
A blog about everything, everywhere, all at once. 'Pun intended'
- Published on
How Snappy evolved from the nextjs-fastapi-colpali template into a vision-first document retrieval system- Published on
Quick, practical notes to run the Vidore benchmark smoothly on a single 32GB GPU: dtype, batch size, and common OOM fixes.- Published on
An end-to-end, page-level Vision RAG template with ColPali-style embeddings, Qdrant multivector retrieval (with optional binary quantization), and MinIO-backed storage — dockerized and API-first.- Published on
A little-script to create a FastAPI server for ColQwen2.5