Gemini 4 tops both public and private coding benchmarks record-high, yet Google stock slumps same week
bookwormengr · x · 2026-10-31
- Artificial Analysis data shows Gemini 4 Carbon ranking first on both public tasks (Terminal-Bench 4.0, 65.2%) and private tasks (FrontierCode 2.0, 40.6%); Gemini 4 Argon is second on public (57.1) but drops to ninth on private (25.7).
- Adam Holter framed it as "Gemini 4 beats Opus 5.5 on public coding tasks and loses to Grok on private ones," prompting a snarky reply from DeepMind's Steven Hansen about being "only capable of benchmaxxing."
- The poster notes the irony: Google achieved record performance of any publicly known LLM in the same week its stock got hammered, urging the board not to let VPs squander the achievement.
Related event: Gemini 4 Carbon Tops Public and Private Coding Benchmarks(3 posts)→
More from Fun
- Claude's orange logo inspires a Halloween costume punchline in AI circle — TejasKumar_ · 2026-10-31
- A "spellular automaton": cellular automaton engineered to spell "Carlos" 12 times — CatAstro_Piyush · 2026-10-31
- David Lieb's reaction to subvocal wristband demo goes viral; silent input seen as step toward telepathy — LinusEkenstam · 2026-10-31
- Halloween AI felt-craft characters: pumpkin, skeleton, and crow in matching denim — sergeantsref · 2026-10-31
- 576GB 'go f*ck yourself Sam Altman' mirror spreads as OpenAI docs leak — retr0jirachi · 2026-10-31
- Google's AI Overview keeps giving wrong ages for low-profile musicians — ShowCharming3635 · 2026-10-31