alt Hacker NewsThere’s a lot of debate about whether this reduces model accuracy, but this is basically Chinese grammar and Chinese vibe coding seems to work fine while (supposedly) using 30-40% less tokens
There’s a lot of debate about whether this reduces model accuracy, but this is basically Chinese grammar and Chinese vibe coding seems to work fine while (supposedly) using 30-40% less tokens
It's like googling... if you have skillz/experience you can google almost anything with 3-4 words...