Users report GPT quality drop; an old-school prompt warm-up trick to regain compliance
NickPassig · x · 2026-10-31
NickPassig complains that model quality has degraded across the board: GPT 5.5 wrote great short films, songs and humor last month but now feels watered down, and his NoSpoon agents see instruction-following collapse on gpt 5.6 and 6.1 sol, which he attributes to compute strain. He also shares an old-school trick: prepend a simple warm-up like 'Ready for instructions? Say yes.' before the regular prompt to put the model in a compliant state.
Related event: Users Report Across-the-Board Quality Decline in GPT Models(4 posts)→
More from Models
- Anthropic files for Nasdaq IPO: $16.3B H1 revenue, compute spend down to 0.5x revenue — DeepLearningAI · 2026-10-31
- ThursdAI crew debates: with Opus 7 coming, who still cares about better benchmarks? — thursdai_pod · 2026-10-31
- Swift 1.5 Flash Next Heretic quants add IQ4_XS and IQ3_S tuned for reverse engineering — jjusko20 · 2026-10-31
- 14 models tested on 960 analyst tasks: AI scores 97/100 as analyst, picks winners no better than random — HeyAmit_ · 2026-10-31
- Anthropic adds invisible text watermark to Claude outputs starting Nov 3 — petesena · 2026-10-31
- GPT-6 Luna vs GPT-6.1 Terra vs GPT-6.1 Sol vs GPT-6 Astra in Pixel Art — Angaisb_ · 2026-10-31