FlashDreams is a high-performance inference and serving library for interactive autoregressive video and world models. It began as the optimized runtime behind the NVIDIA OmniDreams closed-loop demo for GTC 2026 and has grown into a general platform for real-time world-model applications across gaming, autonomous vehicles, robotics, simulated or virtual environments, and more.
flashdreams-promo-hq-0601-720P.mp4
- NVIDIA GPU with 80 GB VRAM or more (e.g. H100 80GB), see notes below.
- NVIDIA driver from the R580 series or newer (compatible with CUDA 13.x)
- CUDA 13.x (PyTorch
2.11.0+cu130and thenvidia-*-cu13libraries are resolved byuv sync. A system CUDA toolkit is needed only for the developer extras and is included innvidia/cuda:13.2.1-cudnn-devel-ubuntu24.04) - Python >= 3.10
- PyTorch >= 2.11.0+cu130 (
>= 2.9for bare PyPI library install) - Linux x86-64 or arm64
- 100 GB+ free storage space recommended for environment and model checkpoints.
- Docker with the NVIDIA Container Toolkit (optional, only for the container workflow)
Development and testing were performed on GPUs with 80 GB of VRAM or more. Inference can fail (out-of-memory) on consumer and even enthusiast GPUs. Per-model GPU and VRAM requirements are listed on each model page in the model gallery.
The complete setup is in
the installation guide.
Assuming uv is installed, the shortest viable path is:
git clone https://github.com/NVIDIA/flashdreams.git
cd flashdreams
uv sync --extra runners
export HF_TOKEN=<your-hf-token>
uv run flashdreams-run --helpNote for developers/maintainers you would want to run uv sync --extra dev --extra runners instead.
Then launch your first model by following the quickstart guide. For example, the offline Self-Forcing T2V quickstart command is:
uv run --project integrations/self_forcing \
flashdreams-run self-forcing-wan2.1-t2v-1.3b \
--total-blocks 7You can also install FlashDreams as a library from PyPI:
pip install flashdreamsDrive a world model in real time with the OmniDreams interactive-drive demo. See the
interactive demo guide.
FlashDreams ships first-party integrations under
integrations/. Each model has a dedicated docs page with
runner slugs, multi-GPU commands, and (where available) profiling benchmarks.
| Model | Family |
|---|---|
| Self-Forcing | Streaming Wan2.1 T2V |
| OmniDreams | HDMap-conditioned driving world model |
| LingBot-World | Camera-controllable I2V world model |
| Wan2.1 | Bidirectional T2V / I2V |
| Causal-Forcing | Streaming Wan2.1 T2V / I2V |
| Causal Wan2.2 | FastVideo Causal Wan 2.2 14B MoE T2V |
| FlashVSR | Streaming video super-resolution |
| Cosmos-Predict2.5 | Bidirectional T2V / I2V |
See the model gallery and the new method guide to add your own.
For day-to-day development:
uv sync --extra dev --extra runners
uv run pre-commit run -a
uv run pytest -m "not manual"See DEV.md for repository-specific workflow notes.
For how to contribute, see CONTRIBUTING.md.
New integrations, bug reports, feature requests, performance tuning, and
documentation edits are all welcome.
Use GitHub Issues to report defects or request improvements.
Join us on the NVIDIA Omniverse Discord
to share your results and take part in technical discussion! Channel: #flashdreams
To report a potential security vulnerability, follow the coordinated
disclosure process in SECURITY.md.
FlashDreams is released under the Apache License 2.0. Third-party
components and their licenses are listed in
THIRD-PARTY-NOTICES and NOTICE. The
repository is REUSE-compliant; see REUSE.toml and
LICENSES/.
If FlashDreams is useful in your research or product, please cite the project:
@misc{flashdreams2026,
title = {FlashDreams: High-performance inference and serving for
interactive autoregressive video and world models},
author = {{FlashDreams Contributors}},
year = {2026},
howpublished = {\url{https://github.com/NVIDIA/flashdreams}},
}
@misc{nvidia2026omnidreams,
title={OmniDreams: Real-Time Generative Closed-Loop Autonomous Vehicle Simulation Built on NVIDIA Cosmos},
author={Basant, Aarti and Kar, Amlan and Paschalidou, Despoina and Garcia Cobo, Guillermo and Turki, Haithem and Ling, Huan and Seo, Jaewoo and Wang, Jialiang and Lucas, James and Wu, Jay and Lorraine, Jonathan and Gao, Jun and He, Kai and Tothova, Katarina and Xie, Kevin and Tyszkiewicz, Michal and Wu, Qi and de Lutio, Riccardo and Li, Ruilong and Fidler, Sanja and Kim, Seung Wook and Shen, Tianchang and Cao, Tianshi and Pfaff, Tobias and Lew, William and Ren, Xuanchi and Lu, Yifan and Gojcic, Zan and Wang, Zian},
year={2026},
note={Technical report},
howpublished={\url{https://research.nvidia.com/labs/sil/projects/omnidreams-blog/paper.pdf}}
}