The Weight Is Over - Interactive Diffusion on Consumer GPUs

Abstract:On-device inference is booming, but the momentum is almost all in language models. Diffusion pipelines are memory hungry, latency-sensitive, and require orchestrating an embedder, a transformer, a decoder, and often further postprocessing that is not as standardized as LLM inference loops are. We navigate the trade-off between performance, quality, and model footprint to reach as many client devices in the wild as possible. We make three contributions: an embedding translator that maps a small text encoder into a large encoder space to cut weight and latency; a reproducible sweep recipe for navigating the speed/quality/memory triangle in diffusion pipelines; and an interactive on-device image generation editor achieving sub-second TTFI on recent GPUs.

Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Performance (cs.PF)

arXiv:2609.21849 [cs.LG]

(or arXiv:2609.21849v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2609.21849 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

https://doi.org/10.1145/3829339.3847852 Focus to learn more DOI(s) linking to related resources

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.