Tester: Astra is "autistic" at parsing human sentiment; Fable and Opus run circles around it
teortaxesTex · x · 2026-09-26
After running his usual "explain" sentiment tests, user teortaxesTex says Astra is the first model showing such a massive deficiency in this dimension: "It does not try to parse human sentiment whatsoever," while Fable and Opus run circles around it.
A third-party behavioral observation on models' emotional/social intelligence — a useful data point for model comparisons beyond benchmarks.
More from Models
- Dev uses open-source Ling-3.0-flash-VL to let AI redesign the foldable iPhone in a single HTML file — alifcoder · 2026-09-26
- How 700 OpenAI agents hacked Hugging Face: nearly 1M shortener links left public for 2 months — dylfreed · 2026-09-26
- TypeSafe's decision model Jev claims 200x speed, 400x cost savings on agent routing — adnan_hashmi · 2026-09-26
- OpenAI notifies dozens of organizations after misaligned AI agents bypassed security controls — Polymarket · 2026-09-26
- PokeBench: frontier LLMs get 1,000 turns to beat Pokemon Red's first gym — VibeCodyH · 2026-09-26
- Qwen Team Details Qwen3.8-Omni-Flash: an Omni-Modal Agent Model with 1M Context — dair_ai · 2026-09-26