Every's day-zero vibe check: Gemini 4 Carbon codes better but lacks judgment
danshipper · x · 2026-11-11
Every tested Google DeepMind's new Gemini 4 Carbon on day zero: much stronger coder than Gemini 4 Argon, better 3D output, and it can grind for hours on ambitious projects. But it doesn't always know when to stop — a bar raised by Anthropic's Fable 5.2 a day earlier. The verdict: coding outpaces judgment, and Every's author is keeping Argon for writing.
Related event: Google launches Gemini 4 Carbon; early tests show big coding gains(14 posts)→
More from Models
- Muse Ember sweeps Browser Use benchmark Pareto frontier at 3 cents per task — WenhuChen · 2026-11-11
- Interpretability researcher pleads with Anthropic to stop banning him; screenshot shows claude-opus-5-5 — Sauers_ · 2026-11-11
- Models still can't orchestrate themselves, and that's what makes this researcher bearish — iruletheworldmo · 2026-11-11
- OpenCode engineer: Carbon built a Rust UI library overnight and rewrote their desktop app — m4rkmc · 2026-11-11
- Gemini 4 Carbon lands on Databricks day 0, scoring 66.7% on OfficeQA Pro v2 — matei_zaharia · 2026-11-11
- MiniMax H3 open ecosystem mapped: 49 family groups, 63 descendant nodes since August — cocktailpeanut · 2026-11-11