Newer models need a different prompting style, and old tricks can make outputs worse
emollick · x · 2026-07-21
A reply says using newer, more capable systems requires a step change in how you prompt them—similar to the jump from GPT-4o to GPT-5/o3.
It adds that there is still little accumulated know-how or clear research on prompting these stronger models, and that habits which worked on older models can actually make outputs worse on Fable and Sol.
Related event: Claude Code System Prompt Reduced by 80%(4 posts)→
More from Models
- GPT-6.1 Terra cuts factual error rate from 22.3% to 4.3%, an ~81% drop — haider1 · 2026-10-27
- GPT-6.1 Terra lands: early users praise speed, value and subagent performance — BenBajarin · 2026-10-27
- Leaked GPT-6.1 Terra benchmarks: cheap but 'boring', still far behind Astra and Claude — Angaisb_ · 2026-10-27
- Artificial Analysis Ships Intelligence Index v4.3.2 With GPT-6.1 Terra Comparison — ArtificialAnlys · 2026-10-27
- GPT-6.1 Terra cuts hallucination rate by 50 points while lifting accuracy 2 points — ArtificialAnlys · 2026-10-27
- GPT-6.1 Terra scores 48 on AA Coding Agent Index at one-third the cost of GPT-5.6 Terra — ArtificialAnlys · 2026-10-27