Deorshi Nishant·Oct 1Part 9: Multimodal RAG, When the Answer Is a PictureScans, charts, tables and floor plans, and what really works when the important part of a page is not text
Venkat Subbu·Sep 18Grep the GopuramA few weeks ago, I was trying to open a bank account in India from California. The bank sent me a set of forms that needed to be signed and…
Aditya Mangal·Sep 1I Tested 4 Ways to OCR Images in C# — One Took Just 41 msBy Aditya Mangal — Senior AI Engineer. Every number here was measured first-hand on my own Windows laptop. No cloud APIs, no Python — just…
marchiee·Aug 1Understanding How OCR WorksHave you ever wondered how OCR can read text from an image even when it’s blurry, faded or written in messy handwriting?A response icon1A response icon1
CyberRaya·Jul 24OCR Showdown — Evaluating Three OCR ModelsDeciding which OCR model is the best for day-to-day usage.A response icon1A response icon1
Prince Arora·Jul 22Experiments Benchmarking Local OCR, Vision LLM Models on a Scanned Table In A PDF DocumentOffline benchmark. No cloud. No API keys. Local hardware only.A response icon3A response icon3
Aditya Mangal·Jul 17Docling vs Marker vs MinerU: The Ultimate Open-Source PDF Parser Benchmark (2026) — Which Is Best…The Real Open-Source PDF-to-Markdown Benchmark for 2026 — and Which One Belongs in Your RAG PipelineA response icon1A response icon1
Dyna Joshy·Jul 2Running Gemma-3 OCR Without Cloud DependenciesIf you’ve spent any time building Document AI pipelines, you know the standard playbook: a user uploads an image, you ship it off to a…
Jonas Wacker·Apr 6Open-Weight OCR Models Have Finally Caught Up with Cloud ServicesA hands-on benchmark of 10 OCR methods on real-world documents.A response icon1A response icon1
Mhjorleifsson·Apr 3Local Accurate and Fast OCR with OllamaThe new GLM-OCR model is light, fast, accurate
