HelixWorld is a real-time interactive audio-visual world model from Noiz AI.
Give it an image and a prompt. Walk forward or turn around — picture and sound update together. The spatial field turns with the camera. Audio is not a soundtrack laid on afterwards.
- 2026-09-03. We release HelixWorld Preview v1 inference code and a preview checkpoint on Hugging Face. You can also roam an interactive world in the browser at helixworld.org.
| HelixWorld | |
|---|---|
| Roaming / camera navigation | Yes |
| Joint audio-video | Yes |
| Spatial sound field | Follows viewpoint |
| Interactive demo | helixworld.org |
| Offline inference | Preview v1 |
| Full model, training, and report | Coming soon |
- Linux, Python 3.11, CUDA 12.x
- NVIDIA GPU, 80 GB VRAM recommended
- System
ffmpegandffprobe
A reviewed BF16 single-GPU run uses about 70 GB of VRAM.
git clone https://github.com/NoizAI/HelixWorld.git
cd HelixWorld
conda create -n helixworld-preview python=3.11 -y
conda activate helixworld-preview
python -m pip install -r requirements.txt
python download_models.pymodels/
├── text_encoder/gemma-3-12b/
└── weights/
└── model.safetensors
CUDA_VISIBLE_DEVICES=0 ./run.sh \
--image /path/to/first_frame.png \
--prompt-file examples/prompt.json \
--actions "W:5,right:5,stop:5" \
--perspective first_person \
--num-frames 121 \
--output-dir outputs/demoEdit examples/prompt.json, or pass --video-prompt, --audio-prompt, and --av-prompt. Output is a clean MP4 in <output-dir>/release/native/.
Actions: W A S D left right up down stop. Combine with + (W+D:8). The number after : is latent transitions; 121 frames have 15. The last segment may omit a duration.
@misc{helixworld2026,
title = {HelixWorld: A Real-Time Interactive Audio-Visual World Model},
author = {{Noiz AI}},
year = {2026},
howpublished = {\url{https://github.com/NoizAI/HelixWorld}},
note = {Preview v1 released; technical report, full model, and training code forthcoming}
}Code is Apache 2.0. The weight license is published with the Preview v1 checkpoint.