Skip to content
Get in touchFR
Updated in October 2026
Explained

How much energy does a ChatGPT query use?

Energy per query for ChatGPT, Claude or Gemini is around 0.3 Wh, about the same as a Google search. Depending on the model and the length of the prompt, however, the figure can be 100 times higher.

Guillaume Pakula
By Guillaume Pakula, co-founder of Celsius. Since 2020, he has helped 80+ organisations with their Bilan Carbone® and climate strategy.
June 2026
Updated October 2026 · 8 min
A short text query to ChatGPT uses around 0.3 Wh of energy (Gemini measured at ~0.24 Wh, GPT-4o estimated at ~0.43 Wh before it was withdrawn from ChatGPT in February 2026, GPT-4.1 nano at around 0.2 Wh), about the same as a Google search. On a reasoning model or with a very long prompt, consumption climbs to several tens of Wh. The comparison of the 3 AI assistants shows which one uses the least energy.
Key takeaways
  • 1A short query uses ~0.3 Wh, roughly the energy of a Google search.
  • 2The measurements converge: Gemini at ~0.24 Wh and GPT-4o at ~0.43 Wh (withdrawn from ChatGPT in 2026) fall in the same low range.
  • 3Reasoning or a long prompt raises the figure sharply: up to several tens of Wh.
  • 4The main issue is volume: billions of queries every day.

A typical text query to ChatGPT uses around 0.3 Wh of electricity, or ~0.012 gCO2e if it is processed in France. Every question you type triggers a live calculation, an inference query, the point at which an already trained model produces an answer without looking up a page that has already been written. That calculation runs on specialised processors in a data centre, and it uses that electricity. This article sets out where the figure comes from, why it can be 100 times higher and what you can do with it.

ChatGPT queries

What do your ChatGPT queries use over a year?

Choose your number of queries per day and where the data centre's electricity comes from.

Your queries
Data centre electricity

25 / day, regular use, over a year

2.7kWh/year

= 25 queries × 365 days × 0.3 Wh, that is ≈ 113 gCO2e (French electricity)

≈ 274 h

of an LED bulb (10 W) switched on

≈ 0.05%

of a household's annual electricity

≈ 800 m

in a combustion-engine car (French electricity)

Even at 100 queries a day, a year of usage comes to ~11 kWh, 0.2% of a household's electricity.

The country's electricity mix changes the CO2 by a factor of 11.1; the issue lies in the billions of daily queries taken together.

Sources: Google (arXiv 2508.15734), Epoch AI, OpenAI, Ember (Global Electricity Review 2026, AI Impact Database), ADEME Impact CO2; car ~142 gCO2e/km, household ≈ 5 MWh/year; excludes reasoning models (× 50 to 100).

The 0.3 Wh figure is an order of magnitude, not an audited measurement: proprietary model providers such as OpenAI publish almost nothing official. The article therefore works in ranges: how the figure is built, what makes it vary by a factor of 100 and then the global volume that your query adds to. The converter above already gives the annual equivalents of your daily usage.

1How much energy does a short ChatGPT query use?

For a short text query on a mass-market model, published figures converge on about 0.3 Wh. Google measured a median query to its AI assistant (Gemini) (in French) at ~0.24 Wh (arXiv 2508.15734). Sam Altman, the chief executive of OpenAI, put forward ~0.34 Wh for an average ChatGPT query (The Gentle Singularity). An independent benchmark placed GPT-4o, then the mass-market model behind ChatGPT, at around 0.43 Wh ("How Hungry is AI?"). These 3 independent methods land in the same low range, which translates into the following concrete reference points:

Where the figure comes from

3 independent estimates of the energy used by a query

Each point is an estimate of the energy used by a short query, on an axis in watt-hours.

Measurement

Gemini (Google)

The only production measurement published by a provider: median query on the Gemini apps, August 2025.

Statement

ChatGPT (OpenAI)

The figure put forward by Sam Altman in June 2025, with no published methodology to support it.

Benchmark

GPT-4o

Independent benchmark "How Hungry is AI?" (2025), when GPT-4o powered ChatGPT; it was withdrawn from ChatGPT in February 2026. Epoch AI converges on ~0.3 Wh using a different method.

None of these figures has been audited: OpenAI publishes no official measurement, and only Google has quantified its production. The convergence of 3 independent methods, between 0.2 and 0.5 Wh per short query, nonetheless makes the order of magnitude reliable.

Google (arXiv 2508.15734); S. Altman, The Gentle Singularity (June 2025); "How Hungry is AI?" (arXiv 2505.09598); Epoch AI (2025). "Short" query: ~500 tokens generated. Excludes reasoning and long prompts.
  • Energy: ~0.3 Wh per short query, enough to power a 10 W LED bulb for around 2 minutes.
  • Digital equivalent: about as much as a Google search (Epoch AI).
  • In France: ~0.012 gCO2e (electricity mix ~41 gCO2e/kWh in 2025 according to Ember, an energy think tank), less than a metre travelled by car.
  • In the United States: ~0.12 gCO2e (mix ~384 gCO2e/kWh), the same Wh emitting 9 times more than in France; with the global average (~458 gCO2e/kWh), ~0.14 gCO2e.
Portrait of Sam Altman, chief executive of OpenAI
Statement - June 2025
"About 0.34 watt-hours": the most widely cited figure comes from a blog post
Read the original post

The figure everyone cites for the average consumption of a ChatGPT query comes from a blog post by Sam Altman, The Gentle Singularity: 0.34 Wh, "about what an oven would use in a little over one second", and about 0.32 mL of water. It has no published methodology, no detailed scope and no audit: it is a statement, not a measurement. Its one merit is that the order of magnitude agrees with independent estimates.

You obtain the CO2 emitted by a query by multiplying its energy by the carbon intensity of the electricity grid that supplies the data centre. A query processed in a French data centre, supplied by a heavily nuclear electricity mix, emits a fraction of what the same query emits in the United States. To place a digital consumption figure among other carbon orders of magnitude, you therefore need to know where each Wh is consumed. Our AI carbon footprint calculator does this calculation model by model, including the country of the data centre.

Energy is not the only resource at stake: the same Google measurement also reports ~0.26 mL of water per median query, or 5 drops, detailed in the water footprint of an AI query.

2Why does ChatGPT's energy per query vary so much?

The "0.3 Wh" figure applies to a short query on an optimised model. Change the model or lengthen the prompt and the figure can be multiplied by 100. 3 variables explain most of the gap.

Mistral AI infographic: impact of a page of generated text of 400 tokens - GHG emissions, water consumption and mineral resource use
Mistral AI, July 2025
The first full life cycle assessment of a large model
See the Mistral Large 2 life cycle assessment

Mistral is the only major provider to have published a full life cycle assessment of its model, carried out with Carbone 4 and ADEME (the French Agency for Ecological Transition) and reviewed by third parties (Resilio and Hubblo): a 400-token response from Large 2 through Le Chat comes to 1.14 gCO2e, 45 mL of water and 0.16 mg Sb eq, server manufacturing included. Where other providers report energy in watt-hours, Mistral measures the whole life cycle: it is the sector's methodological reference.

The number of tokens

Consumption is counted in tokens, the basic unit that an AI processes: a fragment of a word, equivalent in French to about 0.75 words. The model reads your input tokens, then builds its answer one token at a time, and every token produced uses energy. Energy therefore follows the volume of tokens, especially on the output side: an exchange of 1,000 input tokens and 1,000 output tokens uses about 3 times more than one of 100 input tokens and 300 output tokens. What counts is the combined length of what goes in and what comes out, far more than the number of questions.

The model running behind ChatGPT

"ChatGPT" is an interface behind which OpenAI runs very different models. Since August 2025, ChatGPT has run on the GPT-5 family, with a fast mode and a thinking mode that uses far more tokens. The public figures date from 2025, when GPT-4o ran most of ChatGPT (0.3 to 0.4 Wh), and OpenAI has published none for the current models. A more capable model draws on more parameters, hence more computation per token and more energy: for the same task, moving from a small model to a large one already multiplies consumption by 10 to 50. It is like choosing an engine: nobody takes the V8 out to fetch a loaf of bread. The detailed comparison by model puts figures on the gap brand by brand.

Prompt length and reasoning

2 situations make consumption rise sharply. The first is long prompts: attaching a 200-page document (~100,000 tokens) pushes a query towards ~40 Wh, whatever the model. The second is reasoning models: before answering, they write an internal monologue (also known as thinking tokens) that you do not see, a long run of tokens in which they work through their reasoning step by step. This hidden draft often amounts to 3 to 15 times as many tokens as the final answer, which takes a query to more than 10 Wh, and up to 29 Wh for DeepSeek-R1 on a long prompt (independent benchmark).

Energy per query

Energy used by prompt length and model

Compare the energy of a short, medium and long query by model type.

Small model (Haiku, Flash)Standard model (Sonnet)Reasoning model
Short prompt~100 tokens, a sentence or a simple question0.05 Wh0.40 Wh3.0 Wh
Medium prompt~1,000 tokens, a long email or a summary0.20 Wh2.5 Wh15 Wh
Long prompt~10,000 tokens, a report or code0.60 Wh5.0 Wh29 Wh

Between a short prompt and a long prompt, energy use is multiplied by 10 to 15. Shortening an instruction or removing an unnecessary attachment matters as much as switching models.

Illustrative orders of magnitude. Method: energy ≈ tokens × consumption per token (see Hugging Face AIEnergyScore, Mistral 2025). For educational use: margin ±30%. Square area is proportional to energy.

Between a short query and this extreme case, the factor exceeds 100, even though the action, typing a question, looks the same. A "per query" figure is therefore meaningless without its context. That is how the myth that a ChatGPT query uses 10 times more energy than a Google search arose, built on a calculation that assumed responses 4 times too long.

3Why does the global volume of AI queries matter so much?

A single query uses little energy, but billions are sent every day, and it is the total that counts. The following 2 facts show the scale.

Orders of magnitude

An AI query compared with everyday actions

0.3 Wh is hard to picture: a one-minute electric shower is worth more than 500 short ChatGPT queries.

AI queriesEveryday actions
Longest query (75,000 words) 40 WhLong query (7,500 words) 2.5 WhShort ChatGPT query 0.3 WhElectric shower (1 min) 160 WhKettle (1.6 L) 100 WhTelevision (1 h) 90 WhSmartphone charge 20 WhMicrowave (1 min) 17 WhTelevision (10 min) 10 WhLaptop (5 min) 2 WhLED bulb (5 min) 0.8 Wh

On its own, an AI query is the same order of magnitude as a trivial everyday action; the energy question comes down to the number of queries made worldwide.

Our World in Data (Hannah Ritchie), based on Epoch AI 2025, inference energy only. A short ChatGPT query is taken as the reference, at 0.3 Wh. Logarithmic scale.
Energy scale

Energy of a query compared with all of generative AI

Click a scale to place it on the axis, from a single query to the whole world.

gap:40,000 billion

1 ChatGPT query

0.34 Wh. One question put to ChatGPT (OpenAI figure, June 2025).

That is the energy of a phone charging for 1 minute.

Between a single query and the global consumption of generative AI, the gap is of the order of 40,000 billion. Measuring AI at the "per query" scale is like measuring road traffic by looking at a single journey.

OpenAI (Sam Altman, June 2025); Schneider Electric (generative AI: 15 TWh in 2025, 347 TWh in 2030); Epoch AI. Conversions: 1 French household ≈ 5 MWh/year. Logarithmic scale: each tick is 1,000 times the previous one.

Inference now outweighs training

First, inference, in other words the use of models, now dominates training. Training is a one-off cost; inference is repeated with every query, across hundreds of millions of users. Every new mass-market use (assistants, agents, AI-augmented search) therefore adds a permanent energy load.

Multiplied by billions of queries every day, that 0.3 Wh drop in the ocean becomes one of the drivers of the doubling of data centre electricity consumption that the International Energy Agency (IEA) forecasts by 2030.

Data centres at ~3% of global electricity by 2030

Second, volume is growing steeply. The number of tokens processed worldwide is projected to rise 24 times between 2026 and 2030 (Goldman Sachs). According to the IEA, Energy and AI, data centre electricity consumption (in French) is set to more than double in 6 years, from ~415 TWh in 2024 to ~945 TWh in 2030. This total represents ~3% of global electricity, a little more than Japan's. That consumption has already grown 4 times faster than total electricity demand since 2017, to the point that, according to the IEA, 1 in 5 data centre projects is at risk of being delayed for lack of a grid connection.

Rows of illuminated server racks in a data centre: the infrastructure that runs billions of AI queries every day
415 TWh in 2024, ~945 TWh in 2030 (IEA): data centre electricity more than doubles, driven by inference, which is repeated with every query.

That is why you need to think in terms of volume when you include AI in a Scope 3 carbon footprint: within an organisation, AI falls within the company's digital emission category, and the figure depends on aggregate volume of use far more than on the individual query.

4How can you reduce the energy your AI queries use?

The 0.3 Wh reference point first puts things in perspective: for an individual, using ChatGPT accounts for a tiny fraction of their footprint, far behind transport or food. In an organisation where use cases are multiplying, a few simple habits cut the energy bill without any loss of quality. The calculator in this section works out the figures for your own texts, model by model.

Calculation method

How much energy and CO2 does your text use?

Choose a model, a hosting country and a text length. The conversion from words → tokens → energy → CO2 is done live.

÷ 0.75

Words

500

× model

Tokens

667

× hosting country

Energy

0.200Wh

CO2e(41 g/kWh)

0.008gCO2e

Moving from France to the United States multiplies the CO2 by ~9. The hosting country matters as much as the choice of model.

Energy factors: independent benchmarks (Mistral AI, Hugging Face AIEnergyScore, 2025-2026), partly on models since withdrawn (GPT-⁠4o left ChatGPT in February 2026). Token factor: ~0.75 word/token (FR/EN). Electricity mix: Ember, Global Electricity Review 2026 (AI Impact Database).
  • Match the model to the task: to summarise, classify, write a draft or answer a simple question, a light model is enough and uses considerably less energy than a large one.
  • Reserve reasoning for hard problems: switching on a reasoning model for an ordinary query wastes energy and money, for a result that is often identical.
  • Keep prompts concise: fewer tokens in and out mechanically means less energy, and there is no need to paste in 10 pages when 2 paragraphs are enough.
  • Measure before you reduce: at company scale, mapping who uses which models is the starting point of a proper digital usage policy, and the input data for the Scope 3 of your carbon footprint.

None of these habits costs anything or degrades the experience. There is no need to feel guilty over every query: the aim is to match the tool to the need, especially when use reaches thousands of queries a month.

5Key takeaways

  • A short ChatGPT query uses ~0.3 Wh (Google reports 0.24 Wh for Gemini and Sam Altman 0.34 Wh), about the energy of a Google search or of an LED left on for 2 minutes.
  • In CO2e terms, expect ~0.012 g in France and ~0.12 g in the United States: the same Wh emits 9 times more depending on the electricity mix that supplies the data centre.
  • The figure varies by a factor of 100 depending on the number of tokens, the size of the model (small or large) and whether reasoning is used, which can push a query to around 30 Wh.
  • These are orders of magnitude: OpenAI publishes almost nothing, only Google has measured a query and the rest comes from independent benchmarks.
  • Volume matters more than your own query: inference now dominates training, and data centre electricity consumption is set to double by 2030, reaching ~3% of global electricity.

The 0.3 Wh figure only makes sense alongside the model, the prompt and the country behind it. For more detail, the comparison of ChatGPT, Claude and Gemini gives the figures model by model. The overview of the carbon footprint of AI puts it all at the scale of an organisation, and the method for counting AI in your carbon footprint turns it into a quantified emission category.

Further resources

Frequently asked questions

or: guillaume@projetcelsius.com

More articles to read

View all →
French environmental cost label for clothing: what changed on 1 October 2026
Regulation

French environmental cost label for clothing: what changed on 1 October 2026

6 min read
What is the French textile eco-score?
Who displays it, and has it become mandatory?
Regulation

What is the French textile eco-score? Who displays it, and has it become mandatory?

12 min read
Removable batteries and battery passports: who is affected in 2027?
Regulation

Removable batteries and battery passports: who is affected in 2027?

14 min read

A senior consultant reads every message

Tell us where you stand and what is pressing, and we will reply within 24 working hours with a first view on how to proceed.

Write to a consultant →
Our toolkit

A full range of free assessment tools

Cost estimator, obligation and eligibility checkers, footprint calculator.

Estimate costs and impacts
Ref. 2026·BC·0184REPORT · FINANCIAL YEAR 2026Bilan Carbone®complete 1·2·3Presentation to the Executive Committee · 2027 action planCONTENTS01Executive summaryp. 402Scope and methodologyp. 1203Emissions by scopep. 24042027 action planp. 5605Methodological appendicesp. 78VOLUME84 pages · ConfidentialSTANDARDISO 14064 · GHG ProtocolQUOTATION · BC.2026.0184Issued 05·05·2026 · Valid for 30 daysCelsius consultancy · Paris 3rd arrondissementINDICATIVE COST ESTIMATEYour companyMid-sized company · 180 employees · CSRD 2027SERVICEBilan Carbone®Scope 1·2·3 · 13 weeksBPIFRANCE FUNDING · - 60%INDICATIVE RANGE EXCL. VAT€16,800to €22,400PROJECT DETAILS01 · SCOPING3 weeks~€4,20002 · MODELLING8 weeks~€9,80003 · PRESENTATION2 weeks~€4,800Indicative estimate · Refined after scopingSIRET 891 234 567 00012

Bilan Carbone® cost estimator

Your price range in 1 minute, based on 2026 market rates and public funding.

1 minStart
Estimate costs and impacts
AI FOOTPRINT · MONTHLY USEYOUR TOTAL FOOTPRINT4.2 kgCO₂eq for 1,000 typical prompts · May 2026MODEL COMPARISON · 1,000 TYPICAL PROMPTSGPT-4oOpenAI6.8 kgClaude OpusAnthropic2.1 kgGemini ProGoogle4.5 kgMistral LargeMistral AI · FR1.6 kg

AI carbon footprint calculator

The climate impact of your AI queries, by model and task. Ecologits methodology.

3 minStart
Check an obligation
DIAGDÉCARBON'ACTIONACCELERATE THETRANSFORMATIONOF YOUR COMPANYIN ENERGY ANDTHE ENVIRONMENTOFFICIAL SCHEME · BPIFRANCE × ADEMECHECKER · ELIGIBILITYELIGIBLEYou meet the 3 Diag Décarbon'Action criteria.YOUR FUNDING€10,000 excl. VAT€6,000excl. VATAmount payable by you after the Bpifrance subsidyFUNDING BREAKDOWNBPIFRANCE 40%YOU 60%€4,000€6,000SCHEME SUBSIDISED BY

Diag Décarbon'Action eligibility

Check in 30 seconds whether your business is eligible for Bpifrance funding covering 40% of a Bilan Carbone® assessment.

30 secStart
Check an obligation
FRENCH REPUBLICMinistry for the Ecological TransitionLEGAL REQUIREMENTGreenhouse GasEmissions AssessmentArticle L. 229-25 of the French Environmental CodeDecree 2022-982 · published 1 July 2022FREQUENCY · 4 YEARSFILING · ADEMECHECKER · BEGES REQUIREMENTREQUIRED720 employees · BEGES requirementPOTENTIAL PENALTY€50,000maximum fine for failing to complete itart. L. 229-25OFFICIAL REFERENCE

BEGES checker

Does the BEGES requirement apply to you? An immediate answer, with the deadline and penalty.

30 secStart
Check an obligation
9:42SCANSCANNING DPP...DPP IDENTIFIEDOrganic cotton T-shirtSKU TX-CB-220 · Batch L-26-04781ESPR COMPLIANT42 DATA POINTS · 5 SECTIONSORIGINIndia · GOTSOrganic cottonFOOTPRINT5.2 kg CO₂eqLCA ISO 14040RECYCLABILITY85%Cotton recycling schemeREPAIRABILITY7.5 / 10Accessible partsEND OF LIFETextile recycling scheme (TLC) · Bordeaux/FRRefashion-approved recovery operatorEU · ESPR 2024/178112.05.2026

DPP checker

Is your product covered by the Digital Product Passport?

1 minStart
Check an obligation
COSTENVIRONMENTAL386POINTS257PER 100 GOfficial methodEcobalyse, v7.0.0CHECKER · ENVIRONMENTAL LABELLINGELIGIBLETextiles · voluntary labellingENVIRONMENTAL COST386 ptsT-shirt 150 g · 257 pts per 100 gExample: Ministry for the Ecological Transition FAQ

Textile environmental labelling 2026

Voluntary labelling, but third parties can publish it from October 2026: where do you stand?

1 minStart