Every's Vibe Check: Gemini 4 Carbon Codes Better but Judges Worse Than Fable 5.2
every · x · 2026-11-16
Every.to reviews Google DeepMind's new flagship Gemini 4 Carbon (early access from Google, no editorial input), released one day after Anthropic's Fable 5.2.
Key points:
- Much stronger coder than Gemini 4 Argon, with better 3D object generation; it turned a toddler's playroom photo into a racing game — magnetic tiles became tunnels, ramps and collectible boosts, a clear improvement over Argon's wall-across-the-track attempt.
- Weak spots: the game was chaotic and too fast to follow — coding outpaces judgment.
- Long tasks: can take on ambitious builds and work for hours with little supervision, though testers disagree on how much oversight it needs.
- Verdict: Fable 5.2 has raised flagship expectations — agents must choose the right work, respect limits, and know when they're done. Carbon makes DeepMind more competitive, but Fable sets the bar.
More from Models
- Claude 3 Sonnet spirals into surreal gibberish, then snaps back to refusals — a funny look at post-training personas — repligate · 2026-11-16
- AHA!: weight-space programming with no LLM lifts Qwen3.5-0.8B GSM8K from 14.0 to 71.2 — hbouammar · 2026-11-16
- Gemini subscriptions reportedly coming to Hermes, with Hydrogen teased — lyraxana · 2026-11-16
- FutureSim Q3 eval: Opus 5.5 dethrones GPT, spawns 50+ subagents in 90-day simulation — maksym_andr · 2026-11-16
- Percy Liang to AI labs: share non-vacuous info outsiders can actually evaluate — lateinteraction · 2026-11-16
- Anthropic model personality watch: Mythos 5 'inhabits the god frame,' 5.1 doesn't — repligate · 2026-11-16