2 ms·
Semantic Search, in the Browser, in a Day
- forgetbook 3mo agoToday I vibe coded a webapp that is simultaneously capable of being hosted for free on github pages, uses a 22mb model (small enough to run in browser) to do semantic search across any 2 pdf's uploaded, and only took 1 working day to build. Are LLM's a bubble?