- 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.'
Vision
All Posts
- generative-ai (13)
- conversational-ai (11)
- nextjs (11)
- rag (10)
- vector-search (7)
- chatbot (7)
- retrieval-augmented-generation (6)
- qdrant (6)
- python (6)
- material-ui (6)
- reactjs (6)
- typescript (6)
- colpali (5)
- openai (5)
- embeddings (5)
vision (4)
- multimodal (4)
- titanium (3)
- fastapi (3)
- oauth (3)
- knowledge-graphs (3)
- llm (3)
- neo4j (3)
- little-scripts (3)
- minio (2)
- frontend (2)
- assistants (2)
- streaming-chat (2)
- ui-ux (2)
- authentication (2)
- computer-vision (2)
- video-processing (2)
- guide (2)
- grafana (2)
- prometheus (2)
- telemetry (2)
- semantic-search (2)
- colqwen (2)
- document-retrieval (2)
- blog (1)
- learning (1)
- ai (1)
- aiethics (1)
- article (1)
- binary-quantization (1)
- gpt-4 (1)
- github (1)
- google (1)
- pinecone-serverless (1)
- unstructured (1)
- nlp (1)
- spark-nlp (1)
- transformers (1)
- nltk (1)
- video-to-text (1)
- object-detection (1)
- rf-detr (1)
- blip-2 (1)
- entity-tracking (1)
- alloy (1)
- mimir (1)
- pyroscope (1)
- loki (1)
- tempo (1)
- opentelemetry (1)
- colbert (1)
- late-interaction (1)
- colnomic (1)
- audio (1)
- gradio (1)
- coplpali (1)
- api-development (1)
- vidore (1)
- benchmark (1)
- gpu-poor (1)
- pytorch (1)
- mteb (1)
- persistent-memory (1)
- nosql (1)
- Published on
How Snappy evolved from the nextjs-fastapi-colpali template into a vision-first document retrieval system- 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
In this article, I'll be integrating Vision into the AI chat assistant I've been building in the previous articles. This work will include creating new UI components for the Vision API, as well as creating new API routes to support the new functionality.