Hacker News: throw7381https://news.ycombinator.com/user?id=throw7381Hacker News RSShttps://hnrss.org/hnrss v2.1.1Sun, 05 Apr 2026 12:53:06 +0000<![CDATA[New comment by throw7381 in "Ingesting PDFs and why Gemini 2.0 changes everything"]]>For data extraction from long documents (100k+ tokens) how does structured outputs via providing a json schema compare vs asking one question per field (in natural language)?

Also I've been hearing good things regarding document retrieval about Gemini 1.5 Pro, 2.0 Flash and gemini-exp-1206 (the new 2.0 Pro?), which is the best Gemini model for data extraction from 100k tokens?

How do they compare against Claude Sonnet 3.5 or the OpenAI models, has anyone done any real world tests?

]]>Wed, 05 Feb 2025 23:22:00 +0000https://news.ycombinator.com/item?id=42956806throw7381https://news.ycombinator.com/item?id=42956806https://news.ycombinator.com/item?id=42956806<![CDATA[New comment by throw7381 in "GPT-4o's Memory Breakthrough – Needle in a Needlestack"]]>Anyone has done any benchmarks for RAG yet?

]]>Tue, 14 May 2024 22:15:10 +0000https://news.ycombinator.com/item?id=40360774throw7381https://news.ycombinator.com/item?id=40360774https://news.ycombinator.com/item?id=40360774<![CDATA[New comment by throw7381 in "MuPDF WASM Viewer Demo"]]>Does anyone know the cost of the commercial licence?

]]>Sat, 20 Apr 2024 11:46:10 +0000https://news.ycombinator.com/item?id=40096611throw7381https://news.ycombinator.com/item?id=40096611https://news.ycombinator.com/item?id=40096611