Paper Overview
- Field: Computer Vision
- Authors: Haoyi Zhu, Haozhe Liu, Yuyang Zhao
- Published: 2026-05-16
- arXiv: 2505.08634
- Trained on only ~213K public video clips with metric-scale pose supervision.
- Training completed in 15 days on 64 H100 GPUs.
- Generates each 60-second clip on a single GPU.
- Distilled variant deploys on a single RTX 5090, denoising a 60s 720p clip in 34 seconds with NVFP4 quantization.
Summary
The authors introduce SANA-WM, an efficient 2.6B-parameter open-source world model natively trained for one-minute generation, synthesizing high-fidelity 720p minute-scale videos with precise camera control. SANA-WM achieves visual quality comparable to large-scale industrial baselines such as LingBot-World and HY-WorldPlay, while significantly improving efficiency.
Core Design
1. Hybrid Linear Attention: combines frame-wise Gated DeltaNet (GDN) with softmax attention for memory-efficient long-context modeling. 2. Dual-Branch Camera Control: ensures precise 6-DoF trajectory adherence. 3. Two-Stage Generation Pipeline: applies a long-video refiner to stage-1 outputs, improving quality and consistency across sequences. 4. Robust Annotation Pipeline: extracts accurate metric-scale 6-DoF camera poses from public videos to produce high-quality, spatiotemporally consistent action labels.
Efficiency Highlights
Benchmark Results
On the authors' one-minute world model benchmark, SANA-WM demonstrates stronger action-following accuracy than previous open-source baselines, and achieves comparable visual quality for scalable world models with 36x higher throughput.
--- *Auto-collected on 2026-05-16*