Class-conditional video generation on Benchmark 17x256x256 resolution (test)
210.9gFVDHi-VAE + DiT
Evaluation Results
| Method | Links | |
|---|---|---|
| Hi-VAE + DiTResolution=17x256x2562026.02 | 210.9 | |
| One-DVAResolution=17x256x256, Method Variant=Full2026.02 | 210.9 | |
| iVideoGPTResolution=17x256x2562026.02 | 254.8 | |
| One-DVAResolution=17x256x256, Method Variant=w/o decoder fine-tuning2026.02 | 274.2 | |
| One-DVAResolution=17x256x256, Method Variant=0% 1D2026.02 | 325.8 | |
| LVDMResolution=17x256x2562026.02 | 372 | |
| LatteResolution=17x256x2562026.02 | 478 | |
| StyleGAN-VResolution=17x256x2562026.02 | 1,431 | |
| VideoGPTResolution=17x256x2562026.02 | 2,880.6 |