SANA-Video 2.0：混合线性注意力与注意力残差实现高效视频生成

摘要

我们提出SANA-Video 2.0，这是一种在统一架构下以5B和14B规模实例化的混合视频扩散Transformer。该模型专为在单GPU上生成长达720p的高质量视频而设计，在质量上可与全softmax视频DiT相媲美，同时保留了线性注意力在长序列处理上的优势。为了避免全程使用二次注意力，混合线性-softmax注意力机制以3:1的比例将门控线性注意力（实现O(N)主导的混合）与周期性门控softmax锚点相结合，恢复了纯线性注意力所缺失的全秩token交互。为在深度上传播这些更新的表示，块注意力残差（AttnRes）将完成的块摘要路由至后续线性层，实现了锚点特征复用，并将深层有效秩提升约12%。通过从头训练，SANA-Video 2.0直接学习完整的混合机制，而非对预训练模型进行线性化处理，并通过降分辨率代理研究确定了25% softmax为质量与效率的最佳折衷点。采用40步采样时，SANA-Video 2.0在单块H100上以480p分辨率、13.2秒内达到VBench评分84.30，在延迟远低于更大规模softmax视频DiT的情况下保持竞争力。其编译后的DiT前向传播在720p/60秒场景下比匹配的全softmax基线快3.2倍，且该差距随视频时长增加而扩大。此外，全栈Sol-Engine优化（内核融合、缓存和稀疏注意力）将此硬件友好型主干网络进一步加速3.58倍，使得5B流水线在720p/5秒场景下仅需13.06秒，比单块H100上的Wan 2.2-A14B快120倍。总体而言，我们的混合设计以大幅降低的成本恢复了softmax级别的表现力，从而解锁了可扩展的长时、高分辨率视频生成能力。

English

We introduce SANA-Video 2.0, a hybrid video diffusion transformer instantiated at 5B and 14B scales under a unified architecture. Designed to generate high-quality video up to 720p on a single GPU, SANA-Video 2.0 matches full-softmax video DiTs in quality while retaining the favorable long-sequence scaling of linear attention. To avoid quadratic attention throughout, Hybrid Linear-Softmax Attention combines gated linear attention for O(N)-dominated mixing with periodic gated-softmax anchors at a 3:1 ratio, restoring the full-rank token interactions that pure linear attention lacks. To propagate these refreshed representations across depth, Block Attention Residuals (AttnRes) route completed block summaries into later linear layers, enabling anchor-feature reuse and boosting deep-layer effective rank by ~12%. Through from-scratch training, SANA-Video 2.0 learns the complete hybrid directly rather than linearizing pretrained models, with reduced-resolution proxy studies establishing 25% softmax as the optimal quality-efficiency trade-off. With 40-step sampling, SANA-Video 2.0 achieves a VBench score of 84.30 in 13.2s at 480p on a single H100, remaining competitive with far larger softmax video DiTs at a fraction of the latency. Its compiled DiT forward pass is 3.2x faster than a matched full-softmax baseline at 720p/60s, a gap that expands with video duration. Furthermore, full-stack Sol-Engine optimization (kernel fusion, caching, and sparse attention) accelerates this hardware-friendly backbone by a further 3.58x, bringing the 5B pipeline to 13.06s at 720p/5s and making it 120x faster than Wan 2.2-A14B on one H100. Overall, our hybrid design recovers softmax-level expressiveness at substantially reduced cost, unlocking scalable long, high resolution video generation.