AI Dictionary › Prompting
Token Efficiency, in Italian Efficienza Token, is the Grace evaluation dimension that measures how concise and precise a prompt is: how much a text communicates in few words, without repetition or unnecessary preamble.
Token Efficiency describes the ability to build dense, redundancy-free prompts: every word must earn its place by contributing to a better output. Negative signals on token efficiency include greetings ("Hi! Hope you can help me..."), apologies ("Sorry to bother you, but..."), repetitions ("So in summary, to recap what I was saying...") and vague formulations that the AI cannot translate into concrete actions.
In the international prompting community, Token Efficiency is also important for economic reasons: AI model APIs are paid per token consumed. An organization with 100 employees using AI daily can save thousands of euros per year by optimizing token efficiency in their system prompts, with no loss of output quality.
In Grace, Token Efficiency and its Italian name Efficienza Token refer to the same evaluation dimension, whose weight varies from scenario to scenario (5% in the first scenario of the catalog, for example). You can find your score for this dimension in the post-scenario breakdown. It is often the dimension that improves most naturally with practice: after 10-15 scenarios, most Grace users automatically write more concise prompts.
From our network
HSE Genius: AI for Safety Data Sheets
Extract SDS data, H phrases and ECHA compliance checks in seconds, powered by AI.
Visit hsegenius.com →From the Agora Intelligence blog
📱 Download the Android app (beta) iOS coming soon
Say what you mean. Get what you need.
Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.