Speed is all you need: On-device acceleration of large diffusion models via GPU-aware optimizations

The proliferation of large diffusion models for image generation has led to a significant increase in model size and inference workloads. On-device ML inference in mobile environments requires meticulous performance optimization and consideration of trade-offs due to resource constraints. Running inference of large diffusion models (LDMs) on-device, driven by the need for cost efficiency and user privacy, presents even greater challenges due to the substantial memory requirements …

android core diffusion diffusion models engineers environments gpu image image generation inference machine learning ml inference mobile optimization performance software software engineers speed systems