{"ID":2841810,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.11446","arxiv_id":"2511.11446","title":"DiffPro: Joint Timestep and Layer-Wise Precision Optimization for Efficient Diffusion Inference","abstract":"Diffusion models produce high quality images but inference is costly due to many denoising steps and heavy matrix operations. We present DiffPro, a post-training, hardware-faithful framework that works with the exact integer kernels used in deployment and jointly tunes timesteps and per-layer precision in Diffusion Transformers (DiTs) to reduce latency and memory without any training. DiffPro combines three parts: a manifold-aware sensitivity metric to allocate weight bits, dynamic activation quantization to stabilize activations across timesteps, and a budgeted timestep selector guided by teacher-student drift. In experiments DiffPro achieves up to 6.25x model compression, fifty percent fewer timesteps, and 2.8x faster inference with Delta FID \u003c= 10 on standard benchmarks, demonstrating practical efficiency gains. DiffPro unifies step reduction and precision planning into a single budgeted deployable plan for real-time energy-aware diffusion inference.","short_abstract":"Diffusion models produce high quality images but inference is costly due to many denoising steps and heavy matrix operations. We present DiffPro, a post-training, hardware-faithful framework that works with the exact integer kernels used in deployment and jointly tunes timesteps and per-layer precision in Diffusion Tra...","url_abs":"https://arxiv.org/abs/2511.11446","url_pdf":"https://arxiv.org/pdf/2511.11446v1","authors":"[\"Farhana Amin\",\"Sabiha Afroz\",\"Kanchon Gharami\",\"Mona Moghadampanah\",\"Dimitrios S. Nikolopoulos\"]","published":"2025-11-14T16:14:58Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Diffusion Model\",\"Transformer\"]","has_code":false}