Hacker News new | ask | show | jobs
by atesgoral 1037 days ago
Ignoring the additional token count cost of HTML, how do you envision rendering HTML that is spit out by an LLM directly (without any processing) to the user?