NVIDIA Open-Sources LocateAnything-3B for Faster Visual Localization
LocateAnything-3B uses parallel decoding to predict bounding boxes at once, improving detection in dense scenes and supporting UI, OCR and documents.
#computer-vision×
PixelRAG: Open-Source Visual Web Scraping & RAG via Page Screenshots
PixelRAG is an open-source retriever that indexes rendered page screenshots, not HTML. Uses Qwen3-VL embeddings, FAISS, and beats text RAG on QA.
Claude Fable 5: Pricing, Vision Upgrade, Data Retention, Claude Code Access
Anthropic unveils Claude Fable 5/Mythos 5: higher API pricing, major vision gains, 30-day request retention, and limited-time Claude Code access until June 22.