Aquarius: A Family of Industry-Level Video Generation Models for Marketing Scenarios

TL;DR

Aquarius introduces a scalable, high-performance family of industry-level video generation models tailored for marketing, enabling efficient, high-fidelity video synthesis across various scenarios with advanced infrastructure and multi-aspect capabilities.

Contribution

The paper presents Aquarius, a comprehensive framework with novel architectures and infrastructure for large-scale, high-quality video generation tailored for industrial marketing applications.

Findings

Achieved 36% MFU at large scale with hybrid parallelism.

Implemented 2.35x inference speedup using diffusion cache and attention optimization.

Supported multi-aspect ratio, multi-resolution, and multi-duration video generation.

Abstract

This report introduces Aquarius, a family of industry-level video generation models for marketing scenarios designed for thousands-xPU clusters and models with hundreds of billions of parameters. Leveraging efficient engineering architecture and algorithmic innovation, Aquarius demonstrates exceptional performance in high-fidelity, multi-aspect-ratio, and long-duration video synthesis. By disclosing the framework's design details, we aim to demystify industrial-scale video generation systems and catalyze advancements in the generative video community. The Aquarius framework consists of five components: Distributed Graph and Video Data Processing Pipeline: Manages tens of thousands of CPUs and thousands of xPUs via automated task distribution, enabling efficient video data processing. Additionally, we are about to open-source the entire data processing framework named "Aquarius-Datapipe". Model Architectures for Different Scales: Include a Single-DiT architecture for 2B models and a Multimodal-DiT architecture for 13.4B models, supporting multi-aspect ratios, multi-resolution, and multi-duration video generation. High-Performance infrastructure designed for video generation model training: Incorporating hybrid parallelism and fine-grained memory optimization strategies, this infrastructure achieves 36% MFU at large scale. Multi-xPU Parallel Inference Acceleration: Utilizes diffusion cache and attention optimization to achieve a 2.35x inference speedup. Multiple marketing-scenarios applications: Including image-to-video, text-to-video (avatar), video inpainting and video personalization, among others. More downstream applications and multi-dimensional evaluation metrics will be added in the upcoming version updates.