Do “thinking” modes use more energy than normal answers?
The claim: “Thinking modes use far more energy than a normal answer.”
Yes. Hugging Face’s AI Energy Score team measured open models directly. Reasoning models used about 30 times more energy on average than models without reasoning, and the same model used 150 to 700 times more with reasoning switched on. Most of the difference is length: the hidden “thinking” is extra text the model writes. These are open models, not ChatGPT, whose company hasn’t published figures for its thinking modes.
The bigger pictureOne person’s share is small, but the total keeps growing: electricity use by AI-focused data centres grew 50% in 2025, by the IEA’s estimate.
What nobody knows yet
- OpenAI, Google and Anthropic don’t publish energy per answer for their thinking modes.
Sources (4)
- Hugging Face, AI Energy Score v2 (open models, measured) Measured
“reasoning models use, on average, 30 times more energy”
Quote checked on the page, 18 Oct 2026 - Hugging Face, AI Energy Score v2 Measured
“between 150 and 700 times more energy used by the same model with reasoning enabled compared to without.”
Quote checked on the page, 18 Oct 2026 - Hugging Face, AI Energy Score v2 Measured
“This can be explained in large part by the number of output tokens generated by the models themselves.”
Quote checked on the page, 18 Oct 2026 - IEA, Key Questions on Energy and AI (2026) Estimated
“Electricity consumption from AI-focused data centres grew even faster, surging 50% in 2025.”
Quote checked on the page, 3 Oct 2026
Cite this
Hugging Face, AI Energy Score v2 (open models, measured), as quoted in Hum, “Do “thinking” modes use more energy than normal answers?”, https://humearth.org/q/reasoning (quote checked 18 Oct 2026).
Cite the original source first; Hum is where you found it.
What do Wh, kWh and TWh mean?
Wh is a watt-hour: a 1-watt light left on for an hour. 1 kWh = 1,000 Wh, and 1 TWh = 1 billion kWh. Hum compares each number with something everyday so you don’t need the units.
Send this answer
Thinking modes use far more energy: about 30× more on average in Hugging Face’s measurements of open models, and 150 to 700× for the same model with reasoning on.