A new intermediate language called NERD, built for code that LLMs generate and run, has been released. Written in an English-word-based natural syntax, it is said to cut token counts by up to roughly 67% versus existing languages, though its developer frames it as an experimental weekend project.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.