Search papers, labs, and topics across Lattice.
This paper introduces a bidirectional framework for estimating the energy consumption of text-to-video (T2V) and text-to-video-audio (T2VA) models based on architectural principles and observable generation parameters, without needing access to model weights or sizes. The authors validate their approach across six open-source models, demonstrating that energy consumption adheres to theoretically derived scaling laws, with a decomposition into quadratic and linear terms reflecting architectural complexity. Achieving below 3% mean absolute percentage error (MAPE) across various architectures, this work provides a robust method for sustainability benchmarking in video generation models.
Energy consumption in video generation models can be accurately predicted without direct access to model details, revealing a surprising adherence to scaling laws.
We present a bidirectional framework for estimating the energy consumption of text-to-video (T2V) and text-to-video-audio (T2VA) models from architectural first principles and observable generation parameters such as resolution and duration, requiring no access to weights, model size, or implementation details. Forward, it predicts energy from generation parameters and architectural principles; backward, it recovers architectural scaling behavior from observed inference times, with accuracy serving as a criterion for architectural validity. Building on the established compute-bound nature of video diffusion models, we demonstrate that each model's energy profile obeys theoretically derived scaling laws, decomposing into quadratic and linear terms whose coefficients directly reflect the underlying architectural complexity. Validated across six open-source models spanning 8.3B-27B parameters and three GPU configurations, this decomposition achieves below 3% MAPE across all architectures. This approach offers a standardized, empirically and theoretically grounded framework for sustainability benchmarking across T2V models and architectures.
