List your company

Image Reconstruction on ImageNet 50k (val)

0.3rFIDVA-VAE
-0.05642.34934.7557.1607Sep 6, 2024Nov 13, 2024Jan 20, 2025Mar 30, 2025Jun 6, 2025Aug 13, 2025Oct 21, 2025

Evaluation results

MethodLinks
VA-VAE#Images=160M2025.100.3———213.6
Open-MAGVIT2-I-In1kToken Type=2D, Tokens=32 x 32, Ratio=8, Train Resolution=128 x 128, Codebook Size=2621442024.090.3427.02100——
UniTokRes=256, Code Shape=16 × 16, Codebook Size=16,384 × 4†2025.030.39————
VFM-VAE#Images=44M2025.100.52———214.1
LlamaGenToken Type=2D, Tokens=32 x 32, Ratio=8, Train Resolution=256 x 256, Codebook Size=163842024.090.5924.45———
SD-VAE#Images=108M2025.100.62———212.1
SemHiTokRes=384, Code Shape=27 × 27, Codebook Size=196,6082025.030.66————
FQGANRes=256, Code Shape=16 × 16 × 2, Codebook Size=16,384 × 2†2025.030.94————
IBQToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=262144, Codebook Dim=2562024.121—0.840.203—
IBQRes=256, Code Shape=16 × 16, Codebook Size=262,1442025.031————
OmiTokenizer-VQToken Type=2D, Tokens=32 x 32, Ratio=8, Train Resolution=256 x 256, Codebook Size=81922024.091.11————
SD-VQGANToken Type=2D, Tokens=32 x 32, Ratio=8, Train Resolution=256 x 256, Codebook Size=163842024.091.14————
SemHiTokRes=256, Code Shape=16 × 16, Codebook Size=196,6082025.031.16————
Open-MAGVIT2-I-In1kToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=2621442024.091.1722.64100——
Open-MAGVIT2Token Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=262144, Codebook Dim=02024.121.17—10.2038—
VQGANToken Type=2D, Tokens=32 x 32, Ratio=8, Train Resolution=256 x 256, Codebook Size=163842024.091.1923.38———
VIT-VQGANToken Type=2D, Tokens=32 x 32, Ratio=8, Train Resolution=256 x 256, Codebook Size=81922024.091.28————
IBQToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=16384, Codebook Dim=2562024.121.37—0.960.2235—
IBQRes=256, Code Shape=16 × 16, Codebook Size=16,3842025.031.37————
TokenFlowRes=256, Code Shape=680, Codebook Size=32,7682025.031.37————
Open-MAGVIT2Token Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=16384, Codebook Dim=02024.121.58—10.2261—
Titok-BToken Type=1D, Tokens=64, Train Resolution=256 x 256, Codebook Size=40962024.091.7————
Titok-BToken Type=1D, Tokens=64, Train Resolution=256 x 256, Codebook Size=4096, Codebook Dim=162024.121.7————
Titok-SToken Type=1D, Tokens=128, Train Resolution=256 x 256, Codebook Size=40962024.091.71————
Titok-SToken Type=1D, Tokens=128, Train Resolution=256 x 256, Codebook Size=4096, Codebook Dim=162024.121.71————
VILA-URes=256, Code Shape=16x16x4‡, Codebook Size=16,3842025.031.8————
LlamaGenToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=163842024.092.1920.7997——
LlamaGenToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=16384, Codebook Dim=82024.122.19—0.970.2281—
LlamaGenRes=256, Code Shape=16 × 16, Codebook Size=16,3842025.032.19————
TokLIPRes=256, Code Shape=16 × 16, Codebook Size=16,3842025.032.19————
Titok-LToken Type=1D, Tokens=32, Train Resolution=256 x 256, Codebook Size=40962024.092.21————
Titok-LToken Type=1D, Tokens=32, Train Resolution=256 x 256, Codebook Size=4096, Codebook Dim=162024.122.21————
SDE(MUSE-VL)Res=256, Code Shape=16 × 16, Codebook Size=32,7682025.032.26————
MaskGITToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=10242024.092.28————
MaskGITToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=1024, Codebook Dim=2562024.122.28————
VQGAN-LCToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=100000, Codebook Dim=82024.122.62—0.990.2212—
VQGAN-LCRes=256, Code Shape=16 × 16, Codebook Size=100,0002025.032.62————
VQGAN-LCToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=16384, Codebook Dim=82024.123.01—0.990.2358—
RQVAERes=256, Code Shape=16 × 16 × 4‡, Codebook Size=16,3842025.033.2————
QLIP-BRes=256, Code Shape=16 × 16, Codebook Size=2^{28}2025.033.21————
VQGAN (reproduced)Token Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=16384, Codebook Dim=2562024.123.98—0.0530.2873—
VQGANToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=16384, Codebook Dim=2562024.124.98—0.0590.2843—
SD-VQGANToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=163842024.095.15————
SD-VQGANToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=16384, Codebook Dim=82024.125.15————
VQGANToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=10242024.097.9419.4———
VQGANToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=1024, Codebook Dim=2562024.127.94—0.44——
LlamaGenToken Type=2D, Tokens=16 x 16, Ratio=16, Train Resolution=256 x 256, Codebook Size=16384, Codebook Dim=2562024.129.21—0.0029——