Thread

Generate fonts where every LLM token is the same width

95 points · 23 comments · z-mach9

totetsu · 1790509068 · focus · HN ↗
This made me realize that my usual practice of typing at 80% accuracy, full of mistakes, when sending input to an llm is probably increasing my input token count.
1. synthos · 1790513935 · focus · HN ↗
  Maybe spell checking for prompt text box would save a few bucks

Unofficial Hacker News client; not affiliated with Y Combinator.