Skip to content

DraftAttention2: Fast Video Diffusion with Low-Resolution-Guided Mixed-Precision Attention

Sep 2026 · 0 citations · 23 references
Computer Science

TL;DR

DraftAttention2 is presented, a training-free framework that uses the low-resolution draft attention map to jointly select attention blocks and assign their numerical precision, and achieves a superior quality-efficiency trade-off over existing efficient video generation methods.

Abstract

Video generation has broad applications in content creation and entertainment. Diffusion transformers have advanced the quality of generated videos, but attention over spatiotemporal tokens becomes increasingly expensive as video resolution and duration increase. We present DraftAttention2, a training-free framework that uses the low-resolution draft attention map to jointly select attention blocks and assign their numerical precision. Specifically, spatial 2D average- and max-pooled queries and keys capture complementary regional statistics to estimate block importance, and a shared ranking assigns higher precision to important blocks, lower precision to less important retained blocks, and skips the rest under configurable budgets. Our analysis separates sparsification error from attention-weighted quantization error, establishing when recovering skipped interactions with low-bit computation tightens the output-error bound. This analysis motivates retaining more interactions at low precision while reserving higher precision for blocks with larger attention mass. To translate these fine-grained assignments into practical speedups, we further develop fused operand preparation and a single attention kernel with precision-specific phases, sharing data movement, softmax statistics, and output accumulation across precisions. Experiments demonstrate that our method achieves a superior quality-efficiency trade-off over existing efficient video generation methods. Notably, its advantage is particularly pronounced for few-step video diffusion, where jointly combining sparsity with 4- and 8-bit mixed-precision computation substantially improves generation quality while retaining significant acceleration. Code is available at https://github.com/anemoi-project/anemoi

View source

Similar papers

Preprint Sep 2026

Where Compute Matters: Heterogeneous Attention for Efficient Video Diffusion

Efficient video generation requires reducing the quadratic cost of self-attention over long spatio-temporal token sequences. Existing efficient-attention methods typically apply the same computation pattern to every token, even though denoising difficulty varies substantially across video regions and evolves throughout...

Olga Zatsarynna, Denis Korzhenkov, Juergen Gall et al. · 0 citations
Preprint Oct 2026

Backend-Agnostic Sparse Attention for Fast High-Resolution Visual Generation

Diffusion Transformers (DiTs) have achieved strong performance in image and video generation, but the quadratic complexity of full attention makes high-resolution generation computationally expensive. Window attention offers an efficient alternative, yet existing methods face a practical trade-off: partitioned window a...

Liao Ma, Jia-Yi Song, Yun-Feng Wu et al. · 0 citations
Preprint Sep 2026

VC-Attention: Value Smoothing and Softmax Casting for Low-bit Attention

VC-Attention is proposed, a training-free low-bit attention framework that addresses diffusion Transformers and quantization scale by pairing Value smoothing with a fused probability Cast, and improves fidelity over low-bit baselines.

Xing-Yang Li, Dong-Yun Zou, Shi-Ning Zhang et al. · 2 citations · ⚡1
Preprint Oct 2026

High-Throughput Generative Video Super-Resolution on a Single Consumer GPU

Generative video super-resolution (VSR) can recover visually compelling high-frequency detail by leveraging diffusion priors, but iterative sampling and heavy spatio-temporal coupling make practical deployments difficult. We study a different operating point: practical HD generative VSR on a single high-end consumer GP...

Georgios Panaretou, S. Chatzis · 0 citations
#artificial intelligence Preprint Oct 2026

MC-Sparse: Deconstructing and Closing the Dense-Sparse Attention Gap in Diffusion Transformers

Sparse attention is a primary approach to reducing the latency of diffusion transformers in long-sequence generation tasks, such as video and high-resolution 3D asset generation. However, existing methods can degrade generation quality and fidelity at high sparsity levels. Through controlled oracle comparisons, we trac...

Jia-Rui Chen, Ze-Qiang Lai, Jiang-Shan Wang et al. · 0 citations
Preprint Oct 2026

iCATS: Fast Video Generation via Interaction-Aware Sparse Attention and Timestep-Adaptive Sparsity

Training-free sparse attention offers a practical acceleration solution to Diffusion Transformers (DiTs) via reducing computations without fine-tuning. It typically involves estimating the importance of query-key regions and deriving sparse masks to compute only the important candidates, which inevitably introduces app...

Cheng-Feng Han, Bao-Le Ai, Xian-Lu Bian et al. · 0 citations

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.