4DAnyone on Gradio: Optimized for <32GB VRAM, 8x Faster
_akhaliq · x · 2026-09-02
A developer integrated the 4DAnyone model into a Gradio Space, enabling multiview consistent video generation from a single input. Key achievements include optimizing the DiT model to run on <32GB VRAM (down from 50GB) and reducing runtime from 40 minutes to 5 minutes for 16 views on a 5090. The public Space is limited to 6 outputs due to queue times, though local setups can generate up to 48.
More from Apps
- Voice AI latency: Why milliseconds decide human-like calls — liit_upp · 2026-09-02
- Google Paper: 90% of Autonomous AI Research Papers Suffer Hallucinations — mikeflache · 2026-09-02
- Lovable integrates Fable 5.1: 17% better fix capability, 31% cheaper — w_hgm · 2026-09-02
- ChatGPT with Search Should Not Be Redefined as a Search Engine — castrotech · 2026-09-02
- Clone any image style in 3 steps using ChatGPT — EXM7777 · 2026-09-02
- Anthropic Is Back? Fable 5.1 Hands-On: Major Gains in Coding and Writing — every · 2026-09-02