Model reference · open weights

RollingForcing

Video TencentARC Text→video 1 build Open weights 0 dl/mo

RollingForcing is an open-weight video model from TencentARC. RollingForcing (BF16) weighs 17.0 GB; the smallest configuration that runs it is RTX 4090 24 GB.

  • RollingForcing is a text-to-video model developed by TencentARC that generates multi-minute video sequences in real time at 16 FPS on a single GPU.
  • It utilizes a rolling-window strategy and an Attention Sink mechanism to maintain long-term coherence and reduce error accumulation over thousands of frames.
  • The model supports English and is released under the MIT license.

Summary of the TencentARC/RollingForcing model card, 2026-10-01

What it is

Released byTencentARC
Released2025-10-22
VRAM17.0 GB for the weights

What it runs on

Memory and cards for RollingForcing (BF16)

17.0 GBweights, file size
537 MBruntime overhead
CardWeightsMemory
RTX 3060 12 GB … RTX 4060 Ti 16 GBdoes not fit
RTX 3090 24 GBfits23.4 GB
RTX 4090 24 GBfits23.4 GB
RTX 5090 32 GBfits31.0 GB
L40S 48 GBfits44.0 GB
A100 80 GBfits78.2 GB
H100 80 GBfits78.1 GB
RTX PRO 6000 Blackwell 96 GBfits93.8 GB
DGX Spark (GB10) 128 GB unifiedfits107 GB
H200 141 GBfits138 GB
B200 180 GBfits176 GB

From the model card

What TencentARC says about RollingForcing

Read the model card

· · · ·

💡 TL;DR: REAL-TIME streaming generation of MULTI-MINUTE videos

  • 🚀 Real-Time at 16 FPS: Stream high-quality video directly from text on a single GPU.
  • 🎬 Minute-Long Videos: Generate coherent, multi-minute sequences with dramatically reduced drift.
  • ⚙️ Rolling-Window Strategy: Denoise frames together in a rolling window for mutual refinement, breaking the chain of error accumulation.
  • 🧠 Long-Term Memory: The novel Attention Sink anchors your video, preserving global context over thousands of frames.
  • 🥇 State-of-the-Art Performance: Outperforms all comparable open-source models in quality and consistency.

📚 Citation

If you find this codebase useful for your research, please cite our paper and consider giving the repo a ⭐️ on GitHub: https://github.com/TencentARC/RollingForcing

@article{liu2025rolling,
  title={Rolling Forcing: Autoregressive Long Video Diffusion in Real Time},
  author={Liu, Kunhao and Hu, Wenbo and Xu, Jiale and Shan, Ying and Lu, Shijian},
  journal={arXiv preprint arXiv:2509.25161},
  year={2025}
}

Quoted from the model card on Hugging Face. The full card is behind the Hugging Face link above.

How it works

How video models work

Prompt / imagestart pointTemporal diffusionframes over timeVideoMP4 clipA video model generates a sequence of coherent frames from your prompt or a starting image.

Running it yourself

Run it on a rented GPU

Rent a machine by the hour. How to run this model is on its model card.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms