GPT-6.2 Astra ranks 2nd on raw intelligence but last on writing vibes, 5-model comparison shows
iruletheworldmo · x · 2026-11-21
A blogger compared five frontier models on raw intelligence vs. writing/human-vibe quality, with a stark split:
- Raw intelligence (Artificial Analysis v5.0.1): Claude Fable 5.2 leads at 65 ($26.54/task), GPT-6.2 Astra follows at 62 with excellent cost efficiency ($4.00), Gemini 4 Neon at 61 ($6.44).
- Writing & Human-Vibe Index (weighted composite of Arena Chat, steerability, writing, AA-Briefcase): Fable 5.2 tops at 84.0, while Astra collapses to last place at 39.5 — 31 points below Neon — mirroring the author's distaste for how the model speaks.
- Recommendation: unlimited budget → Fable 5.2 Max with agents; constrained budget → Gemini 4 Neon, which combines high intelligence with clear communication.
More from Models
- Bio researcher: cancel your $100/mo Claude sub, Gemini 4 has no biosafety theater — DeryaTR_ · 2026-11-21
- Intel researcher: open-source models are good enough for most AI tasks — JFPuget · 2026-11-21
- Inkling 2 Mini hits 613 tok/s decode and 1M context on one RTX 5090 via Basalt — jesdga95 · 2026-11-21
- Single-Shot Benchmarks Are Dead: Evals Are Now a Compute and Protocol Game — Gauri_the_great · 2026-11-21
- OpenAI releases prompt templates to fix GPT-6.2 Astra and Sol quirks — The Decoder · 2026-11-21
- METR Extends Time-Horizon Benchmark: Opus 5.5 Hits 80% Success at ~20-Hour Tasks — ben_j_todd · 2026-11-21