Burkov's one-line model ranking: 5.5 Haiku far below 6 Luna, on par with 3.8 Flash
burkov · x · 2026-10-17
Andriy Burkov, author of The Hundred-Page Machine Learning Book, posted a one-line ranking: 5.5 Haiku ≪ 6 Luna ≈ Gemini 3.8 Flash, implying Claude 5.5 Haiku is clearly the weakest of the three. The names "6 Luna" and "Gemini 3.8 Flash" are unverified rumors and should be taken with caution.
More from Fun
- Opus 5.5 draws stick figures to nail camera composition, wowing users — justin_hart · 2026-10-17
- Meme: running AI from a cardboard box with a CRT monitor — roculus · 2026-10-17
- Dev hot take: those tweeting most about research ideas aren't the ones doing the work — code_star · 2026-10-17
- "Guys i think you've internally achieved AGI" — the meme reaction — RylanSchaeffer · 2026-10-17
- Spend $200 of your $10k salary on Claude Max yourself — it's positive EV, says engineer — generativist · 2026-10-17
- Will Chen: 'Twitter is the Bloomberg of AI research' — willcb · 2026-10-17