Should LLMs just treat text content as an image? (seangoedecke.com)
recommended by 2 users | 2025 | 167 points by ingve | llm internals | original thread |
2 users recommended this thread. What they said:
krackers Oct 27, 2025 | on: DeepSeek OCR | context

(Just noting that https://news.ycombinator.com/item?id=45652952 and the article therein are also worth reading)

saaaaaam Jan 20, 2026 | on: Show HN: Ocrbase – pdf → .md/.json document OCR and… | context

There was an interesting discussion on here a couple of months back about images vs text, driven by this article: https://www.seangoedecke.com/text-tokens-as-image-tokens/

Discussion is here: https://news.ycombinator.com/item?id=45652952

About | Topics | Past | Comments | Recent | Hindsight | Data | Keyboard shortcuts
Hacker News for Hacker News
Not affiliated with Y Combinator or Hacker News. Comments are by their HN authors. Data: open-index/hacker-news (ODC-BY) and the HN API, as of November 3, 2026.