#frugalcomputing

The "local LLM" idea doesn't scale.

• If all LLM users would move to local LLMs, it would make the problem worse because your local compute is never as efficient as that of a specialised data centre. Both hardware utilisation and performance per Watt would be much worse.
• If the local LLM was really as capable as its cloud variant, the cloud operator would be running the "local" version, and it would do so more energy-efficiently. #FrugalComputing (1/2)

Bought a second-hand #ThinkPad L14 Gen 2a (#AMD Ryzen 5 Pro 5650U). Plenty fast enough for what I do, at about 15% of the price of a current model. Installed #FedoraLinux and, so far, everything works, even the fingerprint reader. Repairable, upgradeable, and a small reminder that we don't always need the latest hardware. #FrugalComputing #RightToRepair

In 2025: Global gas consumption grew by 1.6% Global coal consumption grew by 0.7% Global oil consumption grew by 1.3% [1] GHG emissions grew by 2.3% as a result. [2] Data centres are the fastest growing sources of emissions.[2] #ClimateEmergency #FrugalComputing #NoToAI [1] https://www.energyinst.org/statistical-review [2] https://wedocs.unep.org/items/9f0bf855-2069-42a6-a856-4b389f740c5c

Planning to attend a few of the talks at #limits2026 , particularly interested in the ones focused on solar powered services. https://computingwithinlimits.org/2026/#program I particularly appreciate the timings as it falls into the evening for me which is nice. #solar #solarpower #frugalcomputing #permacomputing

Regarding Altman’s claim that humans consume a lot of energy (*): humans' energy consumption is proportional to their wealth, because wealth production requires energy. Wealth is roughly Pareto distributed, which means very few people own most of it and most people own very little: in 2021, 1.1% of the world's population owned 45.8% of world wealth, and is therefore responsible for almost half of the energy consumption and consequently of greenhouse gas emissions. (1/2) #FrugalComputing

The main reason why we should not use #AI: We need to consume LESS energy! Using AI will never accomplish this. Same for most #cloud applications. We need to reduce! #frugalcomputing #reduce