Every's week-long Gemini 4 vibe check: a breakthrough writing model, strong secondary option
danshipper · x · 2026-10-14
Six people at Every spent a week testing Gemini 4 across coding, writing, design, and knowledge work, and were pleasantly surprised:
- Writing is the standout: DeepMind hired professional writers to stamp out AI slop, and it worked — the author calls it his new favorite writing partner for untangling structure and sharpening metaphors
- Positioning: a breakthrough if your company lives in Google's ecosystem; for frontier hoppers, a strong secondary model for certain tasks
- Overall vibe check positive, contrary to what benchmarks suggested
A quoted post adds that Gemini models have always had surprisingly good teaching and explanation skills.
Related event: Every's Hands-on: Gemini 4 Argon Puts Google Back at the Frontier(7 posts)→
More from Models
- Hands-on with Gemini 4 Argon: real-time wildfire and tectonic viz built in 5 minutes — DavidmComfort · 2026-10-14
- Reddit rumor: Gemini 4 Boron to replace 3.1 Pro (unverified) — Any-Field-4793 · 2026-10-14
- Argon pushes Gemma past 120k tokens/sec on a single A100 with vLLM, a 2.9x speedup — osanseviero · 2026-10-14
- Anthropic merge data suggests new models go internal only ~5 weeks before release — ziv_ravid · 2026-10-14
- Perplexity reportedly prepping PPLX-Decider v1.2 27B decision model update this week — ChrisUniverse · 2026-10-14
- Anthropic bans Claude in China, Russia and 10 more countries under new regions policy — scaling01 · 2026-10-14