The hidden cost of an image: Quantifying the energy consumption of Artificial Intelligence image generation

With the growing adoption of Artificial Intelligence (AI) image generation, in conjunction with the ever-increasing environmental resources demanded by AI, we are urged to answer a fundamental question: What is the environmental impact hidden behind each image we generate? In this research, we present a comprehensive empirical experiment designed to assess the energy consumption of AI image generation through diffusion models. Our experiment compares 17 state-of-the-art image generative models by considering multiple factors that could affect their energy consumption, such as model quantization, image resolution, and prompt length. Additionally, we consider established image quality metrics to study potential trade-offs between energy consumption and generated image quality. Results show that image generation models vary drastically in terms of the energy they consume, with up to a 46x difference. Image resolution affects energy consumption inconsistently: increasing resolution fourfold (from 512 × 512 to 1024 × 1024) yields energy increases ranging from 1.3x to 4.8x across models. U-Net-based models tend to consume less than Transformer-based one. Model quantization instead results to deteriorate the energy efficiency of most models, while prompt length and content have no statistically significant impact. Results show that, albeit not common, an effective trade-off between image generation energy consumption and its quality is possible, with few of the models producing the highest quality images also being among the most energy efficient ones.