German IT magazine #ct tested the usual suspect LLMs, focusing on how they handle German language content and their energy footprint. The results speak for themselves. The French LLM #Mistral delivered impressive results, almost matching the quality of the market leader chatGPT while consuming far less energy. You can download this model and run it on-premises with just four #H100 cards (although they still cost 30,000 EUR each). A comparable Meta #LLama model requires four times this number. The more cards you use, the higher your energy consumption, which leads to a cost difference of 5.40 EUR per 1 mio token for Mistral, to 34.40 EUR for LLama (!) just to cover the electricity bill (and a lower quality than Mistral, at least in German). Read the complete article here (German, paywall).