---
title: "Step-Video-T2V Technical Report — paper by Ma et al. (2025)"
description: "The authors also release Step-Video-T2V-Eval, a new benchmark on which the model reportedly… 56 references, cited by 20 papers here."
authors: ["Guoqing Ma", "Haoyang Huang", "Kun Yan", "Liangyu Chen", "Nan Duan", "Shengming Yin", "Changyi Wan", "Ranchen Ming", "Xiaoniu Song", "Xing Chen", "Yu Zhou", "Deshan Sun", "Deyu Zhou", "Jian Zhou", "Kaijun Tan", "Kang An", "Mei Chen", "Wei Ji", "Qiling Wu", "Wen Sun", "Xin Han", "Yanan Wei", "Zheng Ge", "Aojie Li", "Bin Wang", "Bizhu Huang", "Bo Wang", "Brian Li", "Changxing Miao", "Chen Xu", "Chenfei Wu", "Chenguang Yu", "Dapeng Shi", "Dingyuan Hu", "Enle Liu", "Gang Yu", "Ge Yang", "Guanzhe Huang", "Gulin Yan", "Haiyang Feng", "Hao Nie", "Haonan Jia", "Hanpeng Hu", "Hanqi Chen", "Haolong Yan", "Heng Wang", "Hongcheng Guo", "Huilin Xiong", "Huixin Xiong", "Jiahao Gong", "Jianchang Wu", "Jiaoren Wu", "Jie Wu", "Jie Yang", "Jiashuai Liu", "Jiashuo Li", "Jingyang Zhang", "Junjing Guo", "Junzhe Lin", "Kaixiang Li", "Lei Liu", "Lei Xia", "Liang Zhao", "Liguo Tan", "Liwen Huang", "Liying Shi", "Ming Li", "Mingliang Li", "Muhua Cheng", "Na Wang", "Qiaohui Chen", "Qinglin He", "Qiuyan Liang", "Quan Sun", "Ran Sun", "Rui Wang", "Shaoliang Pang", "Shiliang Yang", "Sitong Liu", "Siqi Liu", "Shuli Gao", "Tiancheng Cao", "Tianyu Wang", "Weipeng Ming", "Wenqing He", "Xu Zhao", "Xuelin Zhang", "Xianfang Zeng", "Xiaojia Liu", "Xuan Yang", "Yaqi Dai", "Yanbo Yu", "Yang Li", "Yineng Deng", "Yingming Wang", "Yilei Wang", "Yuanwei Lu", "Yu Chen", "Yu Luo", "Yuchu Luo", "Yuhe Yin", "Yuheng Feng", "Yuxiang Yang", "Zecheng Tang", "Zekai Zhang", "Zidong Yang", "Binxing Jiao", "Jiansheng Chen", "Jing Li", "Shuchang Zhou", "Xiangyu Zhang", "Xinhao Zhang", "Yibo Zhu", "Heung-Yeung Shum", "Daxin Jiang"]
year: 2025
venue: "arXiv:2502.10248"
source: "https://arxiv.org/abs/2502.10248"
doi: "10.48550/arxiv.2502.10248"
arxiv: "2502.10248"
licence: "arXiv — non-exclusive licence to distribute"
licence_url: "http://arxiv.org/licenses/nonexclusive-distrib/1.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/ma-2025-step-video-t2v-technical/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/ma-2025-step-video-t2v-technical/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2025](https://md.research.elamur.ai/free/en/papers/2025/page/4/) / Step-Video-T2V Technical Report: The Practice, Challenges, and Future of Video Foundation Model

# Step-Video-T2V Technical Report: The Practice, Challenges, and Future of Video Foundation Model

**Authors:** [Guoqing Ma](https://md.research.elamur.ai/free/en/authors/guoqing-ma/), [Haoyang Huang](https://md.research.elamur.ai/free/en/authors/haoyang-huang/), [Kun Yan](https://md.research.elamur.ai/free/en/authors/kun-yan/), [Liangyu Chen](https://md.research.elamur.ai/free/en/authors/liangyu-chen/), [Nan Duan](https://md.research.elamur.ai/free/en/authors/nan-duan/), [Shengming Yin](https://md.research.elamur.ai/free/en/authors/shengming-yin/), [Changyi Wan](https://md.research.elamur.ai/free/en/authors/changyi-wan/), Ranchen Ming, Xiaoniu Song, [Xing Chen](https://md.research.elamur.ai/free/en/authors/xing-chen/), [Yu Zhou](https://md.research.elamur.ai/free/en/authors/yu-zhou/), Deshan Sun, [Deyu Zhou](https://md.research.elamur.ai/free/en/authors/deyu-zhou/), Jian Zhou, Kaijun Tan, Kang An, Mei Chen, Wei Ji, Qiling Wu, Wen Sun, Xin Han, Yanan Wei, Zheng Ge, Aojie Li, [Bin Wang](https://md.research.elamur.ai/free/en/authors/bin-wang/), Bizhu Huang, [Bo Wang](https://md.research.elamur.ai/free/en/authors/bo-wang/), Brian Li, Changxing Miao, [Chen Xu](https://md.research.elamur.ai/free/en/authors/chen-xu/), [Chenfei Wu](https://md.research.elamur.ai/free/en/authors/chenfei-wu/), Chenguang Yu, Dapeng Shi, Dingyuan Hu, Enle Liu, [Gang Yu](https://md.research.elamur.ai/free/en/authors/gang-yu/), Ge Yang, Guanzhe Huang, Gulin Yan, Haiyang Feng, Hao Nie, Haonan Jia, Hanpeng Hu, Hanqi Chen, Haolong Yan, [Heng Wang](https://md.research.elamur.ai/free/en/authors/heng-wang/), Hongcheng Guo, Huilin Xiong, Huixin Xiong, Jiahao Gong, Jianchang Wu, Jiaoren Wu, Jie Wu, [Jie Yang](https://md.research.elamur.ai/free/en/authors/jie-yang/), Jiashuai Liu, Jiashuo Li, Jingyang Zhang, Junjing Guo, Junzhe Lin, Kaixiang Li, [Lei Liu](https://md.research.elamur.ai/free/en/authors/lei-liu/), Lei Xia, [Liang Zhao](https://md.research.elamur.ai/free/en/authors/liang-zhao/), Liguo Tan, Liwen Huang, Liying Shi, [Ming Li](https://md.research.elamur.ai/free/en/authors/ming-li/), Mingliang Li, Muhua Cheng, Na Wang, Qiaohui Chen, Qinglin He, Qiuyan Liang, Quan Sun, Ran Sun, [Rui Wang](https://md.research.elamur.ai/free/en/authors/rui-wang/), Shaoliang Pang, Shiliang Yang, Sitong Liu, Siqi Liu, Shuli Gao, Tiancheng Cao, Tianyu Wang, Weipeng Ming, Wenqing He, [Xu Zhao](https://md.research.elamur.ai/free/en/authors/xu-zhao/), Xuelin Zhang, Xianfang Zeng, Xiaojia Liu, Xuan Yang, Yaqi Dai, Yanbo Yu, [Yang Li](https://md.research.elamur.ai/free/en/authors/yang-li/), Yineng Deng, Yingming Wang, Yilei Wang, Yuanwei Lu, [Yu Chen](https://md.research.elamur.ai/free/en/authors/yu-chen/), Yu Luo, Yuchu Luo, Yuhe Yin, Yuheng Feng, Yuxiang Yang, [Zecheng Tang](https://md.research.elamur.ai/free/en/authors/zecheng-tang/), Zekai Zhang, Zidong Yang, [Binxing Jiao](https://md.research.elamur.ai/free/en/authors/binxing-jiao/), Jiansheng Chen, [Jing Li](https://md.research.elamur.ai/free/en/authors/jing-li/), Shuchang Zhou, [Xiangyu Zhang](https://md.research.elamur.ai/free/en/authors/xiangyu-zhang/), [Xinhao Zhang](https://md.research.elamur.ai/free/en/authors/xinhao-zhang/), Yibo Zhu, [Heung-Yeung Shum](https://md.research.elamur.ai/free/en/authors/heung-yeung-shum/) and [Daxin Jiang](https://md.research.elamur.ai/free/en/authors/daxin-jiang/) · **Year:** 2025 · **Venue:** arXiv:2502.10248 · **Licence:** [arXiv — non-exclusive licence to distribute](http://arxiv.org/licenses/nonexclusive-distrib/1.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2502.10248) · [PDF](https://arxiv.org/pdf/2502.10248) · [DOI 10.48550/arxiv.2502.10248](https://doi.org/10.48550/arxiv.2502.10248) · [arXiv 2502.10248](https://arxiv.org/abs/2502.10248) · [OpenAlex](https://openalex.org/W4407632823)

**Citations:** 1 citations (OpenAlex, 2026-10-14) · 20 in our library, of the 149,439 papers whose bibliographies we read

## Summary
Step-Video-T2V, from the StepFun Step-Video Team, is an open-source text-to-video model built to rival commercial systems such as Sora, Kling and Hailuo. It is a 30B-parameter diffusion Transformer (DiT) with 3D full attention, trained with Flow Matching. The DiT works in the latent space of a purpose-built Video-VAE that compresses video 16x16 spatially and 8x in time, which cuts training cost. Two bilingual text encoders let the model accept prompts in Chinese or English.

Training runs in cascaded stages:
- text-to-image pre-training,
- low-resolution text-to-video pre-training, then refinement at higher resolution,
- SFT on a small, carefully curated set,
- Video-DPO, a video-oriented preference optimization step that reduces artifacts.

The authors also release Step-Video-T2V-Eval, a new benchmark on which the model reportedly reaches state-of-the-art quality against open and closed engines. The report is valuable for its candid discussion of open problems: hallucinating captioners, weak handling of rare concept combinations and complex actions, poor physical plausibility, and the high compute cost of long, high-resolution video.

## Key points
- The model has 30B parameters and generates clips of up to 204 frames at 544x992. Code and weights are published at github.com/stepfun-ai/Step-Video-T2V.
- Video-VAE achieves 16x16 spatial and 8x temporal compression while keeping reconstruction quality high.
- Step-Video-T2V-Eval contains 128 prompts across 11 categories. It is bundled with outputs from leading open-source and commercial generators for comparison.
- The authors report several practical findings:
  - Image pre-training supplies basic visual knowledge.
  - Stable low-resolution video pre-training is key for learning motion.
  - A small, diverse, high-quality SFT set matters more than data volume.
  - Overly concentrated cross-attention causes missing objects, which repeating those objects in the prompt can partly fix.
- The paper distinguishes Level-1 (translational) from Level-2 (predictable) video foundation models. As future directions it proposes hybrids of autoregressive and diffusion models, and RL-based post-training.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 Shelves (2): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 2,869 works

🧭 Topics (1): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works

## Cite

Cite as: Guoqing Ma, Haoyang Huang, Kun Yan, Liangyu Chen, Nan Duan, Shengming Yin, Changyi Wan, Ranchen Ming, Xiaoniu Song, Xing Chen, Yu Zhou, Deshan Sun, Deyu Zhou, Jian Zhou, Kaijun Tan, Kang An, Mei Chen, Wei Ji, Qiling Wu, Wen Sun, Xin Han, Yanan Wei, Zheng Ge, Aojie Li, Bin Wang, Bizhu Huang, Bo Wang, Brian Li, Changxing Miao, Chen Xu, Chenfei Wu, Chenguang Yu, Dapeng Shi, Dingyuan Hu, Enle Liu, Gang Yu, Ge Yang, Guanzhe Huang, Gulin Yan, Haiyang Feng, Hao Nie, Haonan Jia, Hanpeng Hu, Hanqi Chen, Haolong Yan, Heng Wang, Hongcheng Guo, Huilin Xiong, Huixin Xiong, Jiahao Gong, Jianchang Wu, Jiaoren Wu, Jie Wu, Jie Yang, Jiashuai Liu, Jiashuo Li, Jingyang Zhang, Junjing Guo, Junzhe Lin, Kaixiang Li, Lei Liu, Lei Xia, Liang Zhao, Liguo Tan, Liwen Huang, Liying Shi, Ming Li, Mingliang Li, Muhua Cheng, Na Wang, Qiaohui Chen, Qinglin He, Qiuyan Liang, Quan Sun, Ran Sun, Rui Wang, Shaoliang Pang, Shiliang Yang, Sitong Liu, Siqi Liu, Shuli Gao, Tiancheng Cao, Tianyu Wang, Weipeng Ming, Wenqing He, Xu Zhao, Xuelin Zhang, Xianfang Zeng, Xiaojia Liu, Xuan Yang, Yaqi Dai, Yanbo Yu, Yang Li, Yineng Deng, Yingming Wang, Yilei Wang, Yuanwei Lu, Yu Chen, Yu Luo, Yuchu Luo, Yuhe Yin, Yuheng Feng, Yuxiang Yang, Zecheng Tang, Zekai Zhang, Zidong Yang, Binxing Jiao, Jiansheng Chen, Jing Li, Shuchang Zhou, Xiangyu Zhang, Xinhao Zhang, Yibo Zhu, Heung-Yeung Shum and Daxin Jiang. 2025. Step-Video-T2V Technical Report: The Practice, Challenges, and Future of Video Foundation Model. arXiv:2502.10248.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@misc{ma2025step,
  title = {{Step-Video-T2V Technical Report: The Practice, Challenges, and Future of Video Foundation Model}},
  author = {Guoqing Ma, Haoyang Huang, Kun Yan, Liangyu Chen, Nan Duan, Shengming Yin, Changyi Wan, Ranchen Ming, Xiaoniu Song, Xing Chen, Yu Zhou, Deshan Sun, Deyu Zhou, Jian Zhou, Kaijun Tan, Kang An, Mei Chen, Wei Ji, Qiling Wu, Wen Sun, Xin Han, Yanan Wei, Zheng Ge, Aojie Li, Bin Wang, Bizhu Huang, Bo Wang, Brian Li, Changxing Miao, Chen Xu, Chenfei Wu, Chenguang Yu, Dapeng Shi, Dingyuan Hu, Enle Liu, Gang Yu, Ge Yang, Guanzhe Huang, Gulin Yan, Haiyang Feng, Hao Nie, Haonan Jia, Hanpeng Hu, Hanqi Chen, Haolong Yan, Heng Wang, Hongcheng Guo, Huilin Xiong, Huixin Xiong, Jiahao Gong, Jianchang Wu, Jiaoren Wu, Jie Wu, Jie Yang, Jiashuai Liu, Jiashuo Li, Jingyang Zhang, Junjing Guo, Junzhe Lin, Kaixiang Li, Lei Liu, Lei Xia, Liang Zhao, Liguo Tan, Liwen Huang, Liying Shi, Ming Li, Mingliang Li, Muhua Cheng, Na Wang, Qiaohui Chen, Qinglin He, Qiuyan Liang, Quan Sun, Ran Sun, Rui Wang, Shaoliang Pang, Shiliang Yang, Sitong Liu, Siqi Liu, Shuli Gao, Tiancheng Cao, Tianyu Wang, Weipeng Ming, Wenqing He, Xu Zhao, Xuelin Zhang, Xianfang Zeng, Xiaojia Liu, Xuan Yang, Yaqi Dai, Yanbo Yu, Yang Li, Yineng Deng, Yingming Wang, Yilei Wang, Yuanwei Lu, Yu Chen, Yu Luo, Yuchu Luo, Yuhe Yin, Yuheng Feng, Yuxiang Yang, Zecheng Tang, Zekai Zhang, Zidong Yang, Binxing Jiao, Jiansheng Chen, Jing Li, Shuchang Zhou, Xiangyu Zhang, Xinhao Zhang, Yibo Zhu, Heung-Yeung Shum and Daxin Jiang},
  year = {2025},
  doi = {10.48550/arxiv.2502.10248},
  eprint = {2502.10248},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2502.10248}
}
```

- [← previous START: Self-taught Reasoner with Tools](https://md.research.elamur.ai/free/en/papers/li-2025-start-self-taught-reasoner-tools/)
- [next → The Distracting Effect: Understanding Irrelevant Passages in RAG](https://md.research.elamur.ai/free/en/papers/amiraz-2025-distracting-effect-understanding-irrelevant-passages/)
- [↑ back to papers of 2025](https://md.research.elamur.ai/free/en/papers/2025/page/4/)

## References

56 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. OpenAI. Video generation models as world simulators. [https://openai.com/index/video-generation-models-as-world-simulators](https://openai.com/index/video-generation-models-as-world-simulators), 2024. [↗](https://openai.com/index/video-generation-models-as-world-simulators)
2. DeepMind. Veo 2. [https://deepmind.google/technologies/veo/veo-2](https://deepmind.google/technologies/veo/veo-2), 2024. [↗](https://deepmind.google/technologies/veo/veo-2)
3. Kuaishou. Kling. [https://klingai.kuaishou.com](https://klingai.kuaishou.com), 2024. [↗](https://klingai.kuaishou.com)
4. MiniMax. Hailuo. [https://hailuoai.com/video](https://hailuoai.com/video), 2024. [↗](https://hailuoai.com/video)
5. RunwayML. Gen-3 alpha. [https://runwayml.com/research/introducing-gen-3-alpha](https://runwayml.com/research/introducing-gen-3-alpha), 2024. [↗](https://runwayml.com/research/introducing-gen-3-alpha)
6. Weijie Kong, Qi Tian, Zijian Zhang, Rox Min, Zuozhuo Dai, Jin Zhou, Jiangfeng Xiong, Xin Li, Bo Wu, Jianwei Zhang, Kathrina Wu, Qin Lin, Junkun Yuan, Yanxin Long, Aladdin Wang, Andong Wang, Changlin Li, Duojun Huang, Fang Yang, Hao Tan, Hongmei Wang, Jacob Song, Jiawang Bai, Jianbing Wu, Jinbao Xue,… — [HunyuanVideo: A Systematic Framework For Large Video Generative Models](https://md.research.elamur.ai/free/en/papers/kong-2024-hunyuanvideo-systematic-framework-large/)
7. Zhuoyi Yang, Jiayan Teng, Wendi Zheng, Ming Ding, Shiyu Huang, Jiazheng Xu, Yuanming Yang, Wenyi Hong, Xiaohan Zhang, Guanyu Feng, et al. [Cogvideox: Text-to-video diffusion models with an expert transformer](https://md.research.elamur.ai/free/en/papers/yang-2024-cogvideox-text-video-diffusion/). arXiv preprint arXiv:2408.06072, 2024a.
8. Zangwei Zheng, Xiangyu Peng, Tianji Yang, Chenhui Shen, Shenggui Li, Hongxin Liu, Yukun Zhou, Tianyi Li, and Yang You. Open-sora: Democratizing efficient video production for all. March 2024. URL [https://github.com/hpcaitech/Open-Sora](https://github.com/hpcaitech/Open-Sora). [↗](https://github.com/hpcaitech/Open-Sora)
9. Bin Lin, Yunyang Ge, Xinhua Cheng, Zongjian Li, Bin Zhu, Shaodong Wang, Xianyi He, Yang Ye, Shenghai Yuan, Liuhan Chen, et al. Open-sora plan: Open-source large video generation model. arXiv preprint arXiv:2412.00131, 2024. — [arXiv:2412.00131](https://arxiv.org/abs/2412.00131) [find](https://scholar.google.com/scholar?q=Bin+Lin%2C+Yunyang+Ge%2C+Xinhua+Cheng%2C+Zongjian+Li%2C+Bin+Zhu%2C+Shaodong+Wang%2C+Xianyi+He%2C+Yang+Ye%2C+Shenghai+Yuan%2C+Liuhan+Chen%2C+et+al%2E+Open%2Dsora+plan%3A+Open%2Dsource+large+video+generation+model%2E+arXiv+preprint+)
10. Patrick Esser, Sumith Kulal, Andreas Blattmann, Rahim Entezari, Jonas Müller, Harry Saini, Yam Levi, Dominik Lorenz, Axel Sauer, Frederic Boesel, Dustin Podell, Tim Dockhorn, Zion English, Kyle Lacey, Alex Goodwin, Yannik Marek, and Robin Rombach. Scaling rectified flow transformers for high-resolut… — [arXiv:2403.03206](https://arxiv.org/abs/2403.03206) [find](https://scholar.google.com/scholar?q=Patrick+Esser%2C+Sumith+Kulal%2C+Andreas+Blattmann%2C+Rahim+Entezari%2C+Jonas+M%C3%BCller%2C+Harry+Saini%2C+Yam+Levi%2C+Dominik+Lorenz%2C+Axel+Sauer%2C+Frederic+Boesel%2C+Dustin+Podell%2C+Tim+Dockhorn%2C+Zion+English%2C+Kyle+Lacey%2C)
11. William Peebles and Saining Xie. [Scalable diffusion models with transformers](https://md.research.elamur.ai/free/en/papers/peebles-2022-scalable-diffusion-models-transformers/), 2023. URL https://arxiv.org/abs/2212.09748. [↗](https://arxiv.org/abs/2212.09748)
12. Adam Polyak, Amit Zohar, Andrew Brown, Andros Tjandra, Animesh Sinha, Ann Lee, Apoorv Vyas, Bowen Shi, Chih-Yao Ma, Ching-Yao Chuang, David Yan, Dhruv Choudhary, Dingkang Wang, Geet Sethi, Guan Pang, Haoyu Ma, Ishan Misra, Ji Hou, Jialiang Wang, Kiran Jagadeesh, Kunpeng Li, Luxin Zhang, Mannat Singh… — [arXiv:2410.13720](https://arxiv.org/abs/2410.13720) [find](https://scholar.google.com/scholar?q=Adam+Polyak%2C+Amit+Zohar%2C+Andrew+Brown%2C+Andros+Tjandra%2C+Animesh+Sinha%2C+Ann+Lee%2C+Apoorv+Vyas%2C+Bowen+Shi%2C+Chih%2DYao+Ma%2C+Ching%2DYao+Chuang%2C+David+Yan%2C+Dhruv+Choudhary%2C+Dingkang+Wang%2C+Geet+Sethi%2C+Guan+Pang%2C+)
13. Lijun Yu, Jose Lezama, Nitesh Bharadwaj Gundavarapu, Luca Versari, Kihyuk Sohn, David Minnen, Yong Cheng, Agrim Gupta, Xiuye Gu, Alexander G Hauptmann, Boqing Gong, Ming-Hsuan Yang, Irfan Essa, David A Ross, and Lu Jiang. Language model beats diffusion - tokenizer is key to visual generation. In The… [find](https://scholar.google.com/scholar?q=Lijun+Yu%2C+Jose+Lezama%2C+Nitesh+Bharadwaj+Gundavarapu%2C+Luca+Versari%2C+Kihyuk+Sohn%2C+David+Minnen%2C+Yong+Cheng%2C+Agrim+Gupta%2C+Xiuye+Gu%2C+Alexander+G+Hauptmann%2C+Boqing+Gong%2C+Ming%2DHsuan+Yang%2C+Irfan+Essa%2C+David+)
14. Nvidia. Cosmos world foundation model platform for physical ai. arXiv preprint arXiv:2501.03575, 2025. — [arXiv:2501.03575](https://arxiv.org/abs/2501.03575) [find](https://scholar.google.com/scholar?q=Nvidia%2E+Cosmos+world+foundation+model+platform+for+physical+ai%2E+arXiv+preprint+arXiv%3A2501%2E03575%2C+2025%2E+%E2%80%94+%5BarXiv%3A2501%2E03575%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2501%2E03575%29)
15. Zongjian Li, Bin Lin, Yang Ye, Liuhan Chen, Xinhua Cheng, Shenghai Yuan, and Li Yuan. Wf-vae: Enhancing video vae by wavelet-driven energy flow for latent video diffusion model. arXiv preprint arXiv:2411.17459, 2024a. — [arXiv:2411.17459](https://arxiv.org/abs/2411.17459) [find](https://scholar.google.com/scholar?q=Zongjian+Li%2C+Bin+Lin%2C+Yang+Ye%2C+Liuhan+Chen%2C+Xinhua+Cheng%2C+Shenghai+Yuan%2C+and+Li+Yuan%2E+Wf%2Dvae%3A+Enhancing+video+vae+by+wavelet%2Ddriven+energy+flow+for+latent+video+diffusion+model%2E+arXiv+preprint+arXiv%3A2)
16. Junyu Chen, Han Cai, Junsong Chen, Enze Xie, Shang Yang, Haotian Tang, Muyang Li, and Song Han. Deep compression autoencoder for efficient high-resolution diffusion models. In The Thirteenth International Conference on Learning Representations, 2025. URL [https://openreview.net/forum?id=wH8XXUOUZU](https://openreview.net/forum?id=wH8XXUOUZU). [↗](https://openreview.net/forum?id=wH8XXUOUZU)
17. Yaron Lipman, Ricky T. Q. Chen, Heli Ben-Hamu, Maximilian Nickel, and Matt Le. [Flow matching for generative modeling](https://md.research.elamur.ai/free/en/papers/lipman-2022-flow-matching-generative-modeling/), 2023. URL https://arxiv.org/abs/2210.02747. [↗](https://arxiv.org/abs/2210.02747)
18. Zhimin Li, Jianwei Zhang, Qin Lin, Jiangfeng Xiong, Yanxin Long, Xinchi Deng, Yingfang Zhang, Xingchao Liu, Minbin Huang, Zedong Xiao, Dayou Chen, Jiajun He, Jiahao Li, Wenyue Li, Chen Zhang, Rongwei Quan, Jianxiang Lu, Jiabin Huang, Xiaoyan Yuan, Xiaoxiao Zheng, Yixuan Li, Jihong Zhang, Chao Zhang,… — [Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding](https://md.research.elamur.ai/free/en/papers/li-2024-hunyuan-dit-powerful-multi/)
19. Ofir Press, Noah A. Smith, and Mike Lewis. Train short, test long: Attention with linear biases enables input length extrapolation, 2022. URL https://arxiv.org/abs/2108.12409. [↗](https://arxiv.org/abs/2108.12409)
20. Junsong Chen, Jincheng Yu, Chongjian Ge, Lewei Yao, Enze Xie, Yue Wu, Zhongdao Wang, James Kwok, Ping Luo, Huchuan Lu, and Zhenguo Li. Pixart-α: Fast training of diffusion transformer for photorealistic text-to-image synthesis, 2023. URL https://arxiv.org/abs/2310.00426. [↗](https://arxiv.org/abs/2310.00426)
21. Jianlin Su, Yu Lu, Shengfeng Pan, Ahmed Murtadha, Bo Wen, and Yunfeng Liu. [Roformer: Enhanced transformer with rotary position embedding](https://md.research.elamur.ai/free/en/papers/su-2021-roformer-enhanced-transformer-rotary-position/), 2023. URL https://arxiv.org/abs/2104.09864. [↗](https://arxiv.org/abs/2104.09864)
22. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. [Training language models to follow instructions with human feedback](https://md.research.elamur.ai/free/en/papers/ouyang-2022-training-language-models-follow-instructions/). Advances in neural information processing systems, 35:27730–27744, 2022.
23. Paul F Christiano, Jan Leike, Tom Brown, Miljan Martic, Shane Legg, and Dario Amodei. Deep reinforcement learning from human preferences. Advances in neural information processing systems, 30, 2017. [find](https://scholar.google.com/scholar?q=Paul+F+Christiano%2C+Jan+Leike%2C+Tom+Brown%2C+Miljan+Martic%2C+Shane+Legg%2C+and+Dario+Amodei%2E+Deep+reinforcement+learning+from+human+preferences%2E+Advances+in+neural+information+processing+systems%2C+30%2C+2017%2E)
24. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D Manning, Stefano Ermon, and Chelsea Finn. [Direct preference optimization: Your language model is secretly a reward model](https://md.research.elamur.ai/free/en/papers/rafailov-2023-direct-preference-optimization-language-model/). Advances in Neural Information Processing Systems, 36, 2024.
25. Bram Wallace, Meihua Dang, Rafael Rafailov, Linqi Zhou, Aaron Lou, Senthil Purushwalkam, Stefano Ermon, Caiming Xiong, Shafiq Joty, and Nikhil Naik. Diffusion model alignment using direct preference optimization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, p… [find](https://scholar.google.com/scholar?q=Bram+Wallace%2C+Meihua+Dang%2C+Rafael+Rafailov%2C+Linqi+Zhou%2C+Aaron+Lou%2C+Senthil+Purushwalkam%2C+Stefano+Ermon%2C+Caiming+Xiong%2C+Shafiq+Joty%2C+and+Nikhil+Naik%2E+Diffusion+model+alignment+using+direct+preference+o)
26. Kai Yang, Jian Tao, Jiafei Lyu, Chunjiang Ge, Jiaxin Chen, Weihan Shen, Xiaolong Zhu, and Xiu Li. Using human feedback to fine-tune diffusion models without any reward model. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8941–8951, 2024b. [find](https://scholar.google.com/scholar?q=Kai+Yang%2C+Jian+Tao%2C+Jiafei+Lyu%2C+Chunjiang+Ge%2C+Jiaxin+Chen%2C+Weihan+Shen%2C+Xiaolong+Zhu%2C+and+Xiu+Li%2E+Using+human+feedback+to+fine%2Dtune+diffusion+models+without+any+reward+model%2E+In+Proceedings+of+the+IEE)
27. Xingchao Liu, Chengyue Gong, and Qiang Liu. [Flow straight and fast: Learning to generate and transfer data with rectified flow](https://md.research.elamur.ai/free/en/papers/liu-2022-flow-straight-fast-learning-generate/). ArXiv, abs/2209.03003, 2022. URL [https://api.semanticscholar.org/CorpusID:](https://api.semanticscholar.org/CorpusID:) 252111177. [↗](https://api.semanticscholar.org/CorpusID:)
28. Sangyun Lee, Zinan Lin, and Giulia Fanti. Improving the training of rectified flows. arXiv preprint arXiv:2405.20320, 2024. — [arXiv:2405.20320](https://arxiv.org/abs/2405.20320) [find](https://scholar.google.com/scholar?q=Sangyun+Lee%2C+Zinan+Lin%2C+and+Giulia+Fanti%2E+Improving+the+training+of+rectified+flows%2E+arXiv+preprint+arXiv%3A2405%2E20320%2C+2024%2E+%E2%80%94+%5BarXiv%3A2405%2E20320%5D%28https%3A%2F%2Farxiv%2Eorg%2Fabs%2F2405%2E20320%29)
29. Deepak Narayanan, Mohammad Shoeybi, Jared Casper, Patrick LeGresley, Mostofa Patwary, Vijay Korthikanti, Dmitri Vainbrand, Prethvi Kashinkunti, Julie Bernauer, Bryan Catanzaro, et al. Efficient large-scale language model training on gpu clusters using megatron-lm. In Proceedings of the International… [find](https://scholar.google.com/scholar?q=Deepak+Narayanan%2C+Mohammad+Shoeybi%2C+Jared+Casper%2C+Patrick+LeGresley%2C+Mostofa+Patwary%2C+Vijay+Korthikanti%2C+Dmitri+Vainbrand%2C+Prethvi+Kashinkunti%2C+Julie+Bernauer%2C+Bryan+Catanzaro%2C+et+al%2E+Efficient+large%2D)
30. Vijay Anand Korthikanti, Jared Casper, Sangkug Lym, Lawrence McAfee, Michael Andersch, Mohammad Shoeybi, and Bryan Catanzaro. [Reducing activation recomputation in large transformer models](https://md.research.elamur.ai/free/en/papers/korthikanti-2022-reducing-activation-recomputation-large-transformer/). Proceedings of Machine Learning and Systems, 5:341–353, 2023.
31. Hao Liu, Matei Zaharia, and Pieter Abbeel. [Ring attention with blockwise transformers for near-infinite context](https://md.research.elamur.ai/free/en/papers/liu-2023-ring-attention-blockwise-transformers-near/). arXiv preprint arXiv:2310.01889, 2023.
32. Sam Ade Jacobs, Masahiro Tanaka, Chengming Zhang, Minjia Zhang, Shuaiwen Leon Song, Samyam Rajbhandari, and Yuxiong He. [Deepspeed ulysses: System optimizations for enabling training of extreme long sequence transformer models](https://md.research.elamur.ai/free/en/papers/jacobs-2023-deepspeed-ulysses-system-optimizations-enabling/). arXiv preprint arXiv:2309.14509, 2023.
33. Samyam Rajbhandari, Jeff Rasley, Olatunji Ruwase, and Yuxiong He. Zero: Memory optimizations toward training trillion parameter models. In SC20: International Conference for High Performance Computing, Networking, Storage and Analysis, pages 1–16. IEEE, 2020. [find](https://scholar.google.com/scholar?q=Samyam+Rajbhandari%2C+Jeff+Rasley%2C+Olatunji+Ruwase%2C+and+Yuxiong+He%2E+Zero%3A+Memory+optimizations+toward+training+trillion+parameter+models%2E+In+SC20%3A+International+Conference+for+High+Performance+Computing)
34. Zili Zhang, Yinmin Zhong, Ranchen Ming, Hanpeng Hu, Jianjian Sun, Zheng Ge, Yibo Zhu, and Xin Jin. Disttrain: Addressing model and data heterogeneity with disaggregated training for multimodal large language models. arXiv preprint arXiv:2408.04275, 2024. [find](https://scholar.google.com/scholar?q=Zili+Zhang%2C+Yinmin+Zhong%2C+Ranchen+Ming%2C+Hanpeng+Hu%2C+Jianjian+Sun%2C+Zheng+Ge%2C+Yibo+Zhu%2C+and+Xin+Jin%2E+Disttrain%3A+Addressing+model+and+data+heterogeneity+with+disaggregated+training+for+multimodal+large+l)
35. PyTorch. Channels Last Memory Format in PyTorch. PyTorch, [https://pytorch.org/tutorials/intermediate/memory_format_tutorial.html](https://pytorch.org/tutorials/intermediate/memory_format_tutorial.html), 2023. Accessed: Oct 4, 2023. [↗](https://pytorch.org/tutorials/intermediate/memory_format_tutorial.html)
36. Philipp Moritz, Robert Nishihara, Stephanie Wang, Alexey Tumanov, Richard Liaw, Eric Liang, Melih Elibol, Zongheng Yang, William Paul, Michael I. Jordan, and Ion Stoica. Ray: A distributed framework for emerging AI applications. In 13th USENIX Symposium on Operating Systems Design and Implementation… [find](https://scholar.google.com/scholar?q=Philipp+Moritz%2C+Robert+Nishihara%2C+Stephanie+Wang%2C+Alexey+Tumanov%2C+Richard+Liaw%2C+Eric+Liang%2C+Melih+Elibol%2C+Zongheng+Yang%2C+William+Paul%2C+Michael+I%2E+Jordan%2C+and+Ion+Stoica%2E+Ray%3A+A+distributed+framework+f)
37. Pritam Damania, Shen Li, Alban Desmaison, Alisson Azzolini, Brian Vaughan, Edward Yang, Gregory Chanan, Guoqiang Jerry Chen, Hongyi Jia, Howard Huang, et al. Pytorch rpc: Distributed deep learning built on tensor-optimized remote procedure calls. Proceedings of Machine Learning and Systems, 5:219–23… [find](https://scholar.google.com/scholar?q=Pritam+Damania%2C+Shen+Li%2C+Alban+Desmaison%2C+Alisson+Azzolini%2C+Brian+Vaughan%2C+Edward+Yang%2C+Gregory+Chanan%2C+Guoqiang+Jerry+Chen%2C+Hongyi+Jia%2C+Howard+Huang%2C+et+al%2E+Pytorch+rpc%3A+Distributed+deep+learning+bui)
38. Ruoyu Qin, Zheming Li, Weiran He, Mingxing Zhang, Yongwei Wu, Weimin Zheng, and Xinran Xu. Mooncake: A kvcache-centric disaggregated architecture for llm serving, 2024. URL https://arxiv.org/abs/2407.00079. [↗](https://arxiv.org/abs/2407.00079)
39. Meta LlamaTeam. [The llama 3 herd of models](https://md.research.elamur.ai/free/en/papers/grattafiori-2024-llama-3-herd-models/), April 2024. URL [https://ai.meta.com/research/publications/the-llama-3-herd-of-models/](https://ai.meta.com/research/publications/the-llama-3-herd-of-models/). [↗](https://ai.meta.com/research/publications/the-llama-3-herd-of-models/)
40. PySceneDetect Developers. PySceneDetect. PySceneDetect. [https://www.scenedetect.com/](https://www.scenedetect.com/). FFmpeg Developers. FFmpeg. FFmpeg. https://ffmpeg.org/. Tsai-Shien Chen, Aliaksandr Siarohin, Willi Menapace, Ekaterina Deyneka, Hsiang-wei Chao, Byung Eun Jeon, Yuwei Fang, Hsin-Ying Lee, Jian Ren, Ming-Hsuan Yang… [↗](https://www.scenedetect.com/)
41. Christoph Schuhmann, Romain Beaumont, Richard Vencu, Cade Gordon, Ross Wightman, Mehdi Cherti, Theo Coombes, Aarush Katta, Clayton Mullis, Mitchell Wortsman, et al. [Laion-5b: An open large-scale dataset for training next generation image-text models](https://md.research.elamur.ai/free/en/papers/schuhmann-2022-laion-5b-open-large-scale/). Advances in Neural Information Processing Systems…
42. LAION. Clip-based nsfw detector. [https://github.com/LAION-AI/CLIP-based-NSFW-Detector](https://github.com/LAION-AI/CLIP-based-NSFW-Detector), 2021. Accessed: [Insert Access Date]. [↗](https://github.com/LAION-AI/CLIP-based-NSFW-Detector)
43. Mingxing Tan and Quoc V Le. [Efficientnet: Rethinking model scaling for convolutional neural networks](https://md.research.elamur.ai/free/en/papers/tan-2019-efficientnet-rethinking-model-scaling-convolutional/). In Proceedings of the 36th International Conference on Machine Learning, pages 6105–6114. PMLR, 2019.
44. PaddleOCR Contributors. Paddleocr. [https://github.com/PaddlePaddle/PaddleOCR](https://github.com/PaddlePaddle/PaddleOCR), 2023. Accessed: [Insert Access Date]. [↗](https://github.com/PaddlePaddle/PaddleOCR)
45. OpenCV Developers. OpenCV. OpenCV, [https://opencv.org/](https://opencv.org/), 2021. Accessed: August 1, 2023. [↗](https://opencv.org/)
46. José Luis Pech-Pacheco, Gabriel Cristóbal, Jesús Chamorro-Martinez, and Joaquín Fernández-Valdivia. Diatom autofocusing in brightfield microscopy: a comparative study. In Proceedings 15th International Conference on Pattern Recognition. ICPR-2000, volume 3, pages 314–317. [find](https://scholar.google.com/scholar?q=Jos%C3%A9+Luis+Pech%2DPacheco%2C+Gabriel+Crist%C3%B3bal%2C+Jes%C3%BAs+Chamorro%2DMartinez%2C+and+Joaqu%C3%ADn+Fern%C3%A1ndez%2DValdivia%2E+Diatom+autofocusing+in+brightfield+microscopy%3A+a+comparative+study%2E+In+Proceedings+15th+Internationa)
47. James Betker, Gabriel Goh, Li Jing, Tim Brooks, Jianfeng Wang, Linjie Li, Long Ouyang, Juntang Zhuang, Joyce Lee, Yufei Guo, et al. Improving image generation with better captions. Computer Science. [https://cdn.openai](https://cdn.openai). com/papers/dall-e-3. pdf, 2(3):8, 2023. [↗](https://cdn.openai)
48. J MacQueen. Some methods for classification and analysis of multivariate observations. In Proceedings of 5-th Berkeley Symposium on Mathematical Statistics and Probability/University of California Press, 1967. [find](https://scholar.google.com/scholar?q=J+MacQueen%2E+Some+methods+for+classification+and+analysis+of+multivariate+observations%2E+In+Proceedings+of+5%2Dth+Berkeley+Symposium+on+Mathematical+Statistics+and+Probability%2FUniversity+of+California+Pre)
49. Xuejin Yang, Xingsheng Zhu, Han Zhang, Zhenyong Hou, and Rui Wang. Chinese clip: Contrastive vision-language pretraining in chinese. arXiv preprint arXiv:2211.01335, 2022. — [arXiv:2211.01335](https://arxiv.org/abs/2211.01335) [find](https://scholar.google.com/scholar?q=Xuejin+Yang%2C+Xingsheng+Zhu%2C+Han+Zhang%2C+Zhenyong+Hou%2C+and+Rui+Wang%2E+Chinese+clip%3A+Contrastive+vision%2Dlanguage+pretraining+in+chinese%2E+arXiv+preprint+arXiv%3A2211%2E01335%2C+2022%2E+%E2%80%94+%5BarXiv%3A2211%2E01335%5D%28https%3A%2F)
50. Xiaoliang Dai, Ji Hou, Chih-Yao Ma, Sam Tsai, Jialiang Wang, Rui Wang, Peizhao Zhang, Simon Vandenhende, Xiaofang Wang, Abhimanyu Dubey, et al. Emu: Enhancing image generation models using photogenic needles in a haystack. arXiv preprint arXiv:2309.15807, 2023. [find](https://scholar.google.com/scholar?q=Xiaoliang+Dai%2C+Ji+Hou%2C+Chih%2DYao+Ma%2C+Sam+Tsai%2C+Jialiang+Wang%2C+Rui+Wang%2C+Peizhao+Zhang%2C+Simon+Vandenhende%2C+Xiaofang+Wang%2C+Abhimanyu+Dubey%2C+et+al%2E+Emu%3A+Enhancing+image+generation+models+using+photogenic+)
51. Xin Tan, Yuetao Chen, Yimin Jiang, Xing Chen, Kun Yan, Nan Duan, Yibo Zhu, Daxin Jiang, and Hong Xu. DSV: Exploiting Dynamic Sparsity to Accelerate Large-Scale Video DiT Training. arXiv preprint arXiv:2502.07590, 2025. [find](https://scholar.google.com/scholar?q=Xin+Tan%2C+Yuetao+Chen%2C+Yimin+Jiang%2C+Xing+Chen%2C+Kun+Yan%2C+Nan+Duan%2C+Yibo+Zhu%2C+Daxin+Jiang%2C+and+Hong+Xu%2E+DSV%3A+Exploiting+Dynamic+Sparsity+to+Accelerate+Large%2DScale+Video+DiT+Training%2E+arXiv+preprint+arXiv)
52. Boyuan Chen, Diego Marti Monso, Yilun Du, Max Simchowitz, Russ Tedrake, and Vincent Sitzmann. Diffusion forcing: Next-token prediction meets full-sequence diffusion, 2024b. URL https://arxiv.org/abs/2407.01392. — [arXiv:2407.01392](https://arxiv.org/abs/2407.01392) [↗](https://arxiv.org/abs/2407.01392)
53. Yoav HaCohen, Nisan Chiprut, Benny Brazowski, Daniel Shalem, Dudu Moshe, Eitan Richardson, Eran Levin, Guy Shiran, Nir Zabari, Ori Gordon, Poriya Panet, Sapir Weissbuch, Victor Kulikov, Yaki Bitterman, Zeev Melumian, and Ofir Bibi. Ltx-video: Realtime video latent diffusion, 2024. URL [https://arxiv.…](https://arxiv.…) [↗](https://arxiv.…)
54. Deyu Zhou, Quan Sun, Yuang Peng, Kun Yan, Runpei Dong, Duomin Wang, Zheng Ge, Nan Duan, Xiangyu Zhang, Lionel M. Ni, and Heung-Yeung Shum. Taming teacher forcing for masked autoregressive video generation, 2025. URL https://arxiv.org/abs/2501.12389. — [arXiv:2501.12389](https://arxiv.org/abs/2501.12389) [↗](https://arxiv.org/abs/2501.12389)
55. DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, Bing Xue, Bingxuan Wang, Bochao Wu, Bei Feng, Chengda Lu, Chengg… — [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/)
56. Contributors and Acknowledgments We designate core contributors as those who have been involved in the development of Step-Video-T2V throughout its entire process, while contributors are those who worked on the early versions or contributed part-time. All contributors are listed in alphabetical orde… [find](https://scholar.google.com/scholar?q=Contributors+and+Acknowledgments+We+designate+core+contributors+as+those+who+have+been+involved+in+the+development+of+Step%2DVideo%2DT2V+throughout+its+entire+process%2C+while+contributors+are+those+who+wor)

*Updated 1 November 2026*