01Rerank documents and search results by relevance to a given query.
02Extract clean, structured content and capture screenshots from web pages.
03Support parallel execution for efficient content reading and searching across multiple sources.
04Perform comprehensive web, academic (arXiv, SSRN), and image searches.
050 GitHub stars
06Deduplicate semantically unique strings and images using embeddings and submodular optimization.