Users report Chat model update brought obsessive over-complication loops
marcoshsq · reddit · 2026-09-06
A developer refactoring ETL pipelines reports that a recent model update degraded the Chat assistant's behavior: plans that used to stay aligned now spiral into loops of over-complication.
- On the older version (5.6 Sol), an API refactoring task finished smoothly with in-scope optimizations.
- On the new version (6 Astra), a simple historical-data backfill ballooned into a mega-plan full of endless tests and steps; asking it to "step back and refocus" just triggers another over-explaining loop.
- The author compares it to a nerdy friend who explains the entire Lord of the Rings lore when you mention the movies, and asks if others see the same.
More from Models
- Dev shows Astra's unusual habit of verifying its own work, unlike any other model — dkundel · 2026-09-06
- Developer halts open-source work over Codex Business quota limits — cnakazawa · 2026-09-06
- Codex Business users can't buy extra resets as weekly limits hit 60% — cnakazawa · 2026-09-06
- Hands-on: GPT-6 Astra is cheaper and faster, Fable 5.1 wins at coding and planning — DRAGONITECOOLGUY · 2026-09-06
- Unverified: GPT-6 Astra reportedly scores 91.8% on SpatialBench vs 80% human baseline — VraserX · 2026-09-06
- DeepSeek v4 models reportedly 30% off via third-party channel, unverified — matlabulous · 2026-09-06