welcome
Science News

Science News

Technology

Technology

How much energy does your AI prompt use? It depends

Science News
Summary
Nutrition label

82% Informative

Large language models (LLMs) produce up to 50 times more CO emissions than others.

They're deployed in massive data centers across the world.

The second half of the model ’s life cycle is inference, which happens every time a user prompts the model.

The environmental impact of a single query can vary dramatically depending on which data is routed to power the data center.

Using a 70 -parameter reasoning model DeepSeek R1 to answer 600,000 questions would emit as much CO as a round-trip flight from London to New York .

Using AI less frequently during the daytime or summer , when power demand spikes, can also make a difference.