Every's 'field guide' ranks Hydrogen, Luna, Spark, Sonnet — as model names go cryptic

every · x · 2026-11-16

Ethan Mollick joked about how cryptic model names have become — and Every followed up with its own 'Vibe Check' field guide chart. Anthropic's Fable 5.2 sits alone at the top (6 of 7 testers gave it gold), Opus 5.5 is the workhorse; OpenAI's GPT-6 Astra excels at computer use but has 'bad habits,' GPT-6 Sol is the cheap daily driver; Google's Gemini 4 Carbon 'codes for hours, doesn't know when to stop,' Argon is the writer; Meta's Muse models have no Vibe Check yet.

Original post →

More from Fun

Fun channel →