SANA-WM: Efficient 2.6B-Parameter World Model for Minute-Scale 720p Video Generation

Forum topic · 小凯 · 2026-05-16

Summary

SANA-WM is an efficient 2.6B-parameter open-source world model natively trained for one-minute video generation, producing high-fidelity 720p minute-scale videos with precise camera control. It matches the visual quality of large industrial baselines like LingBot-World and HY-WorldPlay while greatly improving efficiency. Four core designs enable this: hybrid linear attention combining frame-wise Gated DeltaNet (GDN) with softmax attention for memory-efficient long-context modeling; dual-branch camera control for accurate 6-DoF trajectory adherence; a two-stage generation pipeline with a long-video refiner; and a robust annotation pipeline extracting metric-scale 6-DoF camera poses from public videos. Trained on only ~213K annotated public video clips over 15 days on 64 H100 GPUs, SANA-WM generates each 60-second clip on a single GPU. Its distilled NVFP4-quantized variant denoises a 60s 720p clip in 34 seconds on one RTX 5090. On a one-minute world model benchmark, it shows stronger action-following accuracy than prior open-source baselines and comparable visual quality at 36x higher throughput.

Introduction

This forum post summarizes the paper SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Attention (arXiv:2505.08634), in the computer vision field, authored by Haoyi Zhu, Haozhe Liu, and Yuyang Zhao.

Paper Overview

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 Contributions

Four core designs drive the architecture:

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

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 runs on a single RTX 5090 with NVFP4 quantization, denoising a 60s 720p clip in 34 seconds

Benchmark Results

On the authors' one-minute world model benchmark, SANA-WM demonstrates stronger action-following accuracy than prior open-source baselines, and achieves visual quality comparable to scalable world models at 36x higher throughput.