# FVAttn: Adaptive Sparse Attention with Runtime Load Balancing for Video Generation

arXiv: https://arxiv.org/abs/2607.16190 · HTML: https://mlsanity.com/paper/2607.16190/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Video generation, one of 14 directions in Diffusion models.
- Builds on 22 of its 27 references that are in the map.
- First posted on arXiv 17 Jul 2026.

## Placed in

- Diffusion models / [Video generation](https://mlsanity.com/diffusion/video-generation/)

## Builds on (most cited first, 22 of 22)

- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [VBench: Comprehensive Benchmark Suite for Video Generative Models](https://mlsanity.com/paper/2311.17982/) (2023)
- [One-step Diffusion with Distribution Matching Distillation](https://mlsanity.com/paper/2311.18828/) (2023)
- [Improved Distribution Matching Distillation for Fast Image Synthesis](https://mlsanity.com/paper/2405.14867/) (2024)
- [Open-Sora: Democratizing Efficient Video Production for All](https://mlsanity.com/paper/2412.20404/) (2024)
- [Ring Attention with Blockwise Transformers for Near-Infinite Context](https://mlsanity.com/paper/2310.01889/) (2023)
- [From Slow Bidirectional to Fast Autoregressive Video Diffusion Models](https://mlsanity.com/paper/2412.07772/) (2024)
- [DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models](https://mlsanity.com/paper/2309.14509/) (2023)
- [Sparse VideoGen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity](https://mlsanity.com/paper/2502.01776/) (2025)
- [Vidu: a Highly Consistent, Dynamic and Skilled Text-to-Video Generator with Diffusion Models](https://mlsanity.com/paper/2405.04233/) (2024)
- [SageAttention: Accurate 8-Bit Attention for Plug-and-play Inference Acceleration](https://mlsanity.com/paper/2410.02367/) (2024)
- [HunyuanVideo 1.5 Technical Report](https://mlsanity.com/paper/2511.18870/) (2025)
- [Fast Video Generation with Sliding Tile Attention](https://mlsanity.com/paper/2502.04507/) (2025)
- [Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutation](https://mlsanity.com/paper/2505.18875/) (2025)
- [USP: A Unified Sequence Parallelism Approach for Long Context Generative AI](https://mlsanity.com/paper/2405.07719/) (2024)
- [SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse-Linear Attention](https://mlsanity.com/paper/2509.24006/) (2025)
- [Training-Free Efficient Video Generation via Dynamic Token Carving](https://mlsanity.com/paper/2505.16864/) (2025)
- [db-SP: Accelerating Sparse Attention for Visual Generative Models with Dual-Balanced Sequence Parallelism](https://mlsanity.com/paper/2511.23113/) (2025)
- [S-HPLB: Efficient LLM Attention Serving via Sparsity-Aware Head Parallelism Load Balance](https://mlsanity.com/paper/2603.10353/) (2026)
- [OSP-Next: Efficient High-Quality Video Generation with Sparse Sequence Parallelism, HiF8 Quantization, and Reinforcement Learning](https://mlsanity.com/paper/2605.28691/) (2026)

## Built on by (most cited first, 1 of 1)

- [SparSP: Exploiting Communication Sparsity for Sequence-Parallel Video DiTs](https://mlsanity.com/paper/2609.32197/) (2026)

## Abstract (arXiv)

Video Diffusion Transformers process long spatio-temporal sequences, making self-attention the main bottleneck in high-resolution video generation. Training-free sparse attention reduces this cost, but adaptive Top-p routing creates uneven per-head workloads under multi-GPU sequence parallelism. The resulting workload heterogeneity turns sparse attention into a rank-level straggler problem. We present method, a training-free sparse-attention system that improves the distributed execution efficiency of adaptive sparse attention under multi-GPU sequence parallelism. method uses Top-p routing, a Top-k safety floor, and video-aware block organization as the sparse-routing frontend, then repairs the materialized mask at runtime. Runtime Load Balancing migrates a small number of heavy heads via P2P communication to shorten the current critical path. Slack-Aware Sparse Augmentation fills residual non-critical-rank slack with additional high-value blocks, while overlap hides scheduling and migration overhead behind existing computation. On step-distilled Wan2.2 I2V, method reduces average load imbalance from 1.34 to 1.08 and delivers a 4.41× attention speedup over FlashAttention, while achieving a 2.02--2.11× DiT inference speedup with competitive video quality.