Open-Sora 2.0: Training a Commercial-Level Video Generation Model in $200k

2025-03-12

AI Summary

To address the prohibitively high training costs and commercial barriers of video generation models, this paper introduces Open-Sora 2.0—the first open-source video generation model achieving production-level performance within a $200K budget. Methodologically, it integrates fine-grained data curation, lightweight spatiotemporal attention, progressive training scheduling, and distributed system optimizations to substantially improve training efficiency and hardware utilization. Experimental results demonstrate that Open-Sora 2.0 matches HunyuanVideo and Runway Gen-3 Alpha in human evaluation and VBench benchmarks. Crucially, all model architectures, training code, and checkpoint weights are fully open-sourced, ensuring complete reproducibility. This work establishes, for the first time, the feasibility of a low-cost, high-fidelity, fully open video generation paradigm—significantly lowering the barrier to practical deployment and advancing the democratization of AI-powered video synthesis.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Generation

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Large language models for search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Video generation models have achieved remarkable progress in the past year. The quality of AI video continues to improve, but at the cost of larger model size, increased data quantity, and greater demand for training compute. In this report, we present Open-Sora 2.0, a commercial-level video generation model trained for only $200k. With this model, we demonstrate that the cost of training a top-performing video generation model is highly controllable. We detail all techniques that contribute to this efficiency breakthrough, including data curation, model architecture, training strategy, and system optimization. According to human evaluation results and VBench scores, Open-Sora 2.0 is comparable to global leading video generation models including the open-source HunyuanVideo and the closed-source Runway Gen-3 Alpha. By making Open-Sora 2.0 fully open-source, we aim to democratize access to advanced video generation technology, fostering broader innovation and creativity in content creation. All resources are publicly available at: https://github.com/hpcaitech/Open-Sora.

Problem

• Reducing training cost for commercial-level video generation models.
• Achieving high-quality video generation with efficient resource utilization.
• Democratizing access to advanced video generation technology.

Innovation

• Cost-effective training at $200k
• Efficient data curation and model architecture
• Open-source commercial-level video generation