{"ID":6023534,"CreatedAt":"2026-07-08T01:00:23.257252134Z","UpdatedAt":"2026-07-10T11:42:49.717029521Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.06173","arxiv_id":"2607.06173","title":"MobileWan: Closing the Quality Gap for Mobile Video Diffusion","abstract":"Recent advances in video diffusion have been driven by scaling transformer-based architectures to billions of parameters, substantially improving visual fidelity and motion coherence. In contrast, existing mobile video diffusion models remain limited to relatively small parameter budgets, typically 0.4-1.8B, restricting generation quality. In this work, we show that high-quality mobile video generation does not require small models. Instead, we demonstrate that a server-scale 5B-parameter video diffusion transformer can be deployed efficiently on memory-constrained mobile hardware through recurrent reformulation and structured compression. Starting from Wan2.2-5B, we rely on a recurrence distillation framework that converts video generation into a chunk-wise autoregressive process with constant-memory attention computation. Combined with causal linear attention, the model operates as an RNN at inference time while preserving temporal coherence across chunks. We further propose a learnable attention head pruning method based on binary per-head gates optimized end-to-end using a noise-biased sparsity objective and distillation-based finetuning. Together with sampling-step distillation and memory-optimized VAE decoding, MobileWan becomes the first 5B-scale video diffusion model deployable on a commercial mobile device. Our system generates 5-second 480x832 videos at 16 FPS in 20 seconds end-to-end latency, achieving a VBench score of 83.79 and establishing a new state of the art in mobile video generation. Project page: https://qualcomm-ai-research.github.io/mobilewan","short_abstract":"Recent advances in video diffusion have been driven by scaling transformer-based architectures to billions of parameters, substantially improving visual fidelity and motion coherence. In contrast, existing mobile video diffusion models remain limited to relatively small parameter budgets, typically 0.4-1.8B, restrictin...","url_abs":"https://arxiv.org/abs/2607.06173","url_pdf":"https://arxiv.org/pdf/2607.06173v1","authors":"[\"Mohsen Ghafoorian\",\"Denis Korzhenkov\",\"Adil Karjauv\",\"Ioannis Lelekas\",\"Noor Fathima\",\"Spyridon Stasis\",\"Hanno Ackermann\",\"Boris van Breugel\",\"Markus Nagel\",\"Fatih Porikli\",\"Animesh Karnewar\",\"Amirhossein Habibian\"]","published":"2026-07-07T11:52:46Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\",\"Transformer\",\"Variational Autoencoder\"]","has_code":false}