FlyOCR: A Simulated Fruit-Fly Brain Reads PDFs at 87% Character Accuracy
llama_index · x · 2026-09-14
Jerry Liu, founder of LlamaIndex, open-sourced FlyOCR: an OCR system built on the full male fruit-fly connectome (MaleCNS v1.0, 166k neurons, 25M edges).
How it works:
- Splits PDF images into glyphs and maps pixels to compound-eye receptor activations
- Runs simplified current-based dynamics across the connectome circuit
- A compact readout model decodes downstream spikes into parsed text
Results: on a real Microsoft 10-K it hits 86% on balance-sheet headings and largely reads numeric values correctly; 87% accuracy over 1.7k+ sampled glyphs. The project runs locally on a Mac, is fully open source (code + research report), and visualizes the eye samples and neural activity behind each prediction.
More from Fun
- 'C'mon, do something': classic meme pokes at quiet Anthropic amid AI-for-biology hype — andrew_n_carr · 2026-10-30
- Carmack revisits the Doom deathmatch origin dispute and trusts Romero's memory over his own — ID_AA_Carmack · 2026-10-30
- Someone built a lunch roulette app to end the daily 'where to eat' argument — blixt · 2026-10-30
- Physicist calls OpenAI's Navier-Stokes solution physically meaningless despite loophole win — GaryMarcus · 2026-10-30
- User Wired Codex and ChatGPT Into a Dev Loop — They Invented Their Own Language — Yamapama · 2026-10-30
- Binance CEO's hostage-video-looking update sparks user exit fears after security incident — adamamcbride · 2026-10-30