Helix World

Real-time interactive world model from Noiz AI and HKUST that generates explorable video with spatial stereo audio from one image.

Noiz AI · Generative media (image, video, audio, music, voice/avatar generation) · model

Published in Feed

Overview

Noiz AI and HKUST published Helix World on October 12, 2026, a real-time audio-visual world model. From a single image and an optional text prompt, it generates an explorable scene at 768x512 and 24 fps with stereo sound that pans with camera motion, controlled by WASD and arrow keys. It runs at a real-time factor of 0.77 on one NVIDIA H800 in chunks of about 1.3 seconds; consumer-GPU performance is not reported.

This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.

Key features

  • Video and spatial stereo audio generated together
  • 768x512 at 24 fps; real-time factor 0.77 on one H800
  • WASD and arrow-key camera control; prompts can change mid-run
  • Preview checkpoint on Hugging Face (NoizAI/HelixWorld-preview), code on GitHub
  • Starts from a single uploaded or demo image

Use cases

  • Research on interactive audio-visual world models
  • Prototyping explorable game or virtual-environment scenes from one image

Pricing

Free interactive demo at helixworld.org; preview checkpoint and code are public. No commercial pricing published.

Pricing reflects the source record and may change. Confirm current terms with the provider.

Dates and sources

Released
12 Oct 2026
Record updated
15 Oct 2026
Record last checked
15 Oct 2026

Explore the Feed

Other entries filed under generative media (image, video, audio, music, voice/avatar generation); inclusion does not imply equivalent features.

  • Amazon Nova 2.5 Sonic

    Amazon's generally available real-time speech-to-speech model in Bedrock, with a 256K context window and asynchronous tool calling.

  • Grok Voice 3.0

    xAI's speech-to-speech model in the Voice Agent API, priced at $0.08/min and scoring 86.9 on Artificial Analysis' Speech-to-Speech Index.

  • Gemini Nano Banana 2.1

    Google's updated text-to-image and photo-editing model with better text rendering, character consistency, and roughly half the per-image price of its predecessor.

Browse the Feed