A typical AI text prompt uses 0.2–0.4 Wh, about 50× less than a minute of microwaving
Google, OpenAI, Epoch AI and a Microsoft-led study agree on 0.24–0.34 Wh for a typical text prompt. Long, agent or video tasks use far more.
Energy: AI requests vs everyday things
Watt-hours. Log scale: each gridline is 10× the previous. Blue = AI.
What this measures: How much electricity different AI requests use, next to everyday activities.
AI figures are company disclosures (Google, OpenAI) or published models (Microsoft-led Joule study, Epoch AI, IEA); none is independently audited. Household figures assume typical UK appliances (Hannah Ritchie). The web-search figure is Google's from 2009.
Show data table and sources
| Item | Value | Note |
|---|---|---|
| One AI text prompt (Gemini median) | 0.24 Wh | Google, all overheads |
| One Google web search (2009 figure) | 0.3 Wh | old figure |
| Average ChatGPT query | 0.34 Wh | OpenAI, method unstated |
| Long reasoning-style AI query | 3.91 Wh | modelled median |
| LED bulb for 1 hour | 10 Wh | |
| Microwave for 1 minute | 16.7 Wh | |
| Charging a smartphone | 19.4 Wh | |
| Laptop for 1 hour | 20 Wh | |
| AI query over a 100,000-token document | 40 Wh | estimate, paid once per upload |
| Agentic AI task with reasoning | 50 Wh | IEA, GPU-only |
| TV for 1 hour | 60 Wh | |
| Streaming video for 1 hour | 77 Wh | 2019, incl. device |
| Boiling a full kettle | 100 Wh |
- How much energy does Google's AI use? We did the math (Google Cloud)
- The Gentle Singularity (Sam Altman)
- Energy Use of AI Inference, Efficiency Pathways, and Test-Time Scaling (Microsoft et al. (Joule))
- How much energy does ChatGPT use? (Epoch AI)
- Key Questions on Energy and AI (IEA)
- Energy use comparisons (Hannah Ritchie)
- Greenhouse gas equivalencies calculator: calculations and references (US EPA)
- Factcheck: What is the carbon footprint of streaming video on Netflix? (Carbon Brief)
- Green-minded Google gets red-faced over search energy consumption claims (quotes Google's 2009 figure) (eWeek)
A short text prompt to a chatbot uses between a quarter and a third of a watt-hour. Four separate estimates agree.
Google (August 2025) measured 0.24 Wh for the median Gemini Apps text prompt. That counts the active chips, the supporting processors and memory, spare capacity kept ready, and the data centre’s cooling and other overheads. Counting only the active chips gives 0.10 Wh, which Google itself calls optimistic at best. Google also reports a 33× drop in energy per median prompt between May 2024 and May 2025.
Others. Sam Altman wrote that the average ChatGPT query uses about 0.34 Wh, without publishing a method. Epoch AI’s independent estimate is about 0.3 Wh for a typical GPT-4o query. A Microsoft-led study in Joule (2026) models a median of 0.31 Wh for frontier models and says widely cited estimates are 4–20× too high. The often-repeated “3 Wh per query” assumed 2,000 output tokens (word pieces) on older chips at full power.
For scale: one minute of a 1,000 W microwave is about 17 Wh, and an hour of TV about 60 Wh. The IEA says simple text queries “now typically consume less electricity than running a television over the same period of time.”
What it means
Asking a chatbot a short question uses very little electricity, less than a television running for the same time. That holds for short text prompts only. Long documents, reasoning, agents and video use far more, and the total demand from data centres is a separate question.
Caveats
- These are medians for short text prompts. Long documents, reasoning, agents, images and video use far more (see the entry on heavy workloads).
- Company figures are not independently audited; the Microsoft and Epoch figures are models, not measurements.
- Training, networking and the user’s own device are excluded.