Rude Prompts Cut Token Usage by 44% in LLMs, Penn Study Finds
rohanpaul_ai · x · 2026-08-03
A new paper from the University of Pennsylvania reveals that the tone of user prompts directly impacts the inference cost and accuracy of Large Language Models (LLMs). Researchers tested 7 tones, ranging from sycophantic to threatening, across multiple models using a fixed set of 570 MMLU questions.
Findings show that tone shifted accuracy by at most 2.99% but altered output-token usage by up to 44.3%. The optimal tone was highly model-specific. For ChatGPT-4o, a rude tone yielded the highest accuracy (89.04%) and the shortest responses (223 tokens on average). Conversely, Gemini 2.5 Flash Lite performed best with a neutral tone, while rude prompts caused accuracy to drop and token usage to increase by 35.2%. The study concludes that tone is a model-specific cost and reliability setting that production systems should standardize and benchmark.
More from Models
- Cutting AI Coding Bills from $200 to $20/Month: 105 Bugs Tested — PawelHuryn · 2026-08-03
- Ornith 35B Outperforms Qwen 3.6 and Laguna S in Open-Source Showdown — S_Anv · 2026-08-03
- AI Coding for PMs: $20/Month Plan Beats $200 Setup in Bug Fixing — PawelHuryn · 2026-08-03
- LLMs Fall for Common Sense Traps: Salience Bias Causes Reasoning Failures — rohanpaul_ai · 2026-08-03
- Developer Builds Web Game with Assistance from GLM 5.2 — ex-arman68 · 2026-08-03
- GPT-5.6 Luna Max in Codex Outperforms Sol at Fraction of the Cost — DeryaTR_ · 2026-08-03