"The University of Rhode Island based its report on its estimates that producing a medium-length, 1,000-token GPT-5 response can consume up to 40 watt-hours (Wh) of electricity, with an average just over 18.35 Wh, up from 2.12 Wh for GPT-4. This was higher than all other tested models, except for OpenAI's o3 (25.35 Wh) and Deepseek's R1 (20.90 Wh)."
These numbers don't pass sanity check for me. With 4x300W cards you can get a 1K token DeepSeek R1 output in about 10 seconds. That's just 3.3Wh right? And that's before you even consider batching.
https://www.tomshardware.com/tech-industry/artificial-intell...
https://app.powerbi.com/view?r=eyJrIjoiZjVmOTI0MmMtY2U2Mi00Z...
https://blog.samaltman.com/the-gentle-singularity