arXiv Computer Vision
Spatio-Temporal Parallelism for Diffusion Model Inference on Heterogeneous Multi-GPU Systems
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv Machine Learning
Jul 15
FlashDiff: Efficient Regional Execution and Scheduling for Diffusion Model Serving
arXiv:2607. 12121v1 Announce Type: cross Abstract: Diffusion models have become the central backbone for modern image, video, and audio generation, but their efficient service remains a challenge.
arXiv Machine Learning
Jun 19
TetriServe: Efficiently Serving Mixed DiT Workloads
arXiv:2510. 01565v4 Announce Type: replace Abstract: Diffusion Transformer (DiT) models excel at generating high-quality images through iterative denoising steps, but serving them under strict Service Level Objectives (SLOs) is challenging due to their high computational cost, particularly at larger resolutions.
arXiv Machine Learning
Jul 10
Tessera: Unlocking Heterogeneous GPUs through Kernel-Granularity Disaggregation
arXiv:2604. 10180v2 Announce Type: replace-cross Abstract: Disaggregation maps parts of an AI workload to different types of GPUs, offering a path to utilize modern heterogeneous GPU clusters.
arXiv Machine Learning
Jul 3
GF-DiT: Scheduling Parallelism for Diffusion Transformer Serving
arXiv:2606. 13501v2 Announce Type: replace-cross Abstract: Diffusion Transformers (DiTs) have become the dominant architecture for image and video generation, creating growing demand for efficient DiT serving.
arXiv Machine Learning
3d ago
Mira: Memory-Efficient MoE Inference Using Adaptive Caching and Predictive Expert Staging
arXiv:2609.38090v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models are a compelling architecture for scaling model capacity, making them especially attractive for deployment on resource-...
arXiv AI
Jul 28
MegaSlide-DiT: Memory-Centric Adaptation and Deformable Local Attention for Efficient Video Diffusion
arXiv:2607. 22696v1 Announce Type: cross Abstract: High-resolution video diffusion models built on Diffusion Transformers (DiTs) deliver strong fidelity but quickly exhaust the memory budget of a single workstation.