by ByteDance / Peking University
14B autoregressive diffusion model generating 60-second videos at real-time speed on a single GPU.
Helios, developed by Peking University, ByteDance, and Canva, is a 14B autoregressive diffusion model that generates 60-second videos at real-time speed on a single GPU.
| Use Case | Quality |
|---|---|
| Long-form video clips | Very Good (unique at 60s) |
| Batch content creation | Excellent (real-time speed) |
| Research/experimentation | Excellent (open) |
| Cinematic quality | Good (Runway better) |
Quick tips from the community about what works with Helios right now.
Sign in to share a tip.
No tips yet. Add a tip for this model.