The model already found titles that score higher. Try one.
Runs a local 7B model with SearXNG, browser crawling and a vector DB on consumer hardware. It searches the web, cites sources, remembers pages, works offline after indexing, and needs no API key.
ForesynWanna keep in touch?
Built this solo over a weekend. Soft-launching before the HN post on Monday. If you scored a draft and the prediction either nailed it or whiffed, I want to know.