Image Reconstruction on ImageNet 256
0.27rFIDVAE
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| VAE#Tokens=40962026.06 | 0.27 | — | — | |
| VA-VAE#Tokens=2562026.06 | 0.28 | — | — | |
| HieraTok#Tokens=2562026.06 | 0.45 | — | — | |
| MAETok#Tokens=1282026.06 | 0.48 | — | — | |
| WAN2.2 VAEResolution=2562026.06 | 0.5 | 28.16 | 87 | |
| VQGAN#Tokens=2562026.06 | 0.59 | — | — | |
| SD-VAE#Tokens=2562026.06 | 0.62 | — | — | |
| SelfBootTok#Tokens=642026.06 | 0.66 | — | — | |
| GigaTok XL-XXL#Tokens=2562026.06 | 0.79 | — | — | |
| RepViTokResolution=2562026.06 | 0.8 | 28.9 | 89 | |
| SoftVQ#Tokens=642026.06 | 0.88 | — | — | |
| ViTokResolution=2562026.06 | 0.96 | 28.65 | 89 | |
| VAR-Tok#Tokens=6802026.06 | 1 | — | — | |
| OneD-piece#Tokens=1 to 2562026.06 | 1.08 | — | — | |
| MAR-H#Tokens=2562026.06 | 1.22 | — | — | |
| FlexTok d18-d28#Tokens=1 to 2562026.06 | 1.45 | — | — | |
| TiTok-B#Tokens=642026.06 | 1.7 | — | — | |
| TiTok-S#Tokens=1282026.06 | 1.71 | — | — | |
| LlamaGen-3B#Tokens=5762026.06 | 2.19 | — | — | |
| MaskGIT#Tokens=2562026.06 | 2.28 | — | — | |
| RQ-VAE#Tokens=2562026.06 | 3.2 | — | — |