List your company

Video Reconstruction on UCF-101

3.11rFVDHYDRA-XTOK (Stage 3)
-125.7076743.81121,613.332,482.8488Nov 22, 2024Feb 25, 2025May 31, 2025Sep 3, 2025Dec 7, 2025Mar 12, 2026Jun 16, 2026

Evaluation results

MethodLinks
HYDRA-XTOK (Stage 3)Compression Spatial (fs)=16×, Compression Temporal (ft)=4×, Input Resolution=256×2562026.063.1136.88—0.967—
Wan2.2Compression Spatial (fs)=16×, Compression Temporal (ft)=4×, Input Resolution=256×2562026.064.1536.11—0.961—
AToken-So/C (Stage 3)Compression Spatial (fs)=16×, Compression Temporal (ft)=4×, Input Resolution=256×2562026.067.7734.66—0.953—
MAGVIT-v2-MLMGenerator #Params=307M2026.038.6———1,280
MAGVIT-v2-ARGenerator #Params=840M2026.038.6———1,280
EVATokTokenizer #Params=145M, Generator #Params=327M2026.039.7———774
EVATokTokenizer #Params=145M, Generator #Params=633M2026.039.7———774
TivTok-T512#Tokens=512, #Dim.=32, T/P (%)=0.0492026.0612.6530.260.05330.8982—
Cosmos-M#Tokens=2048, #Dim.=16, T/P (%)=0.1952026.0613.6731.70.05750.9177—
Omni#Tokens=4096, #Dim.=8, T/P (%)=0.3912026.0614.5329.340.04870.925—
TivTok-T1024#Tokens=1024, #Dim.=16, T/P (%)=0.0982026.0617.9729.540.06070.8897—
LARP#Tokens=1024, Codebook Size=8192, Resolution=128x128, Frames=16, Extra Training=additional 75 epochs2026.012028.710.075——
LARP-L-LongTokenizer #Params=173M, Generator #Params=343M2026.0320———1,024
LARP-L-LongTokenizer #Params=173M, Generator #Params=632M2026.0320———1,024
TivTok-T128#Tokens=128, #Dim.=128, T/P (%)=0.0122026.0621.2930.130.06140.901—
STAT-Video#Tokens=1297, Codebook Size=8192, Resolution=128x128, Frames=16, Extra Tokens=3002026.012229.40.059——
STAT-Video#Tokens=1197, Codebook Size=8192, Resolution=128x128, Frames=16, Extra Tokens=2002026.012329.270.06——
TivTok-T512Video resolution=32×256×256, #Tokens=640, #Dim.=322026.0623.2630.250.05910.8948—
SDXL-VAE#Tokens=16384, #Dim.=4, T/P (%)=1.5632026.0623.68————
LARP#Tokens=1024, #Dim.=16, T/P (%)=0.391, Video resolution=16x128x1282026.0623.9328.650.04250.9003—
LARP#Tokens=1024, Codebook Size=8192, Resolution=128x128, Frames=162026.0124————
MAGVIT-MLMTokenizer #Params=158M, Generator #Params=306M2026.0325———1,024
MAGVIT-ARTokenizer #Params=158M, Generator #Params=306M2026.0325———1,024
STAT-Video#Tokens=1097, Codebook Size=8192, Resolution=128x128, Frames=16, Extra Tokens=1002026.012629.080.062——
Omni-DV#Tokens=4096, #Dim.=8, T/P (%)=0.3912026.0627.1228.060.06370.9095—
AdapTok#Tokens=2048, #Dim.=16, T/P (%)=0.7812026.0627.9726.380.05990.8539—
STAT-Video#Tokens=997, Codebook Size=8192, Resolution=128x128, Frames=162026.012928.770.064——
AdapTok#Tokens=1024, Codebook Size=8192, Resolution=128x128, Frames=162026.013625.720.114——
AdapTokTokenizer #Params=195M, Generator #Params=633M2026.0336———1,024
OmniTokenizerCompression Spatial (fs)=8×, Compression Temporal (ft)=4×, Input Resolution=256×2562026.0638.1529.2—0.931—
TivTok-T128Video resolution=32×256×256, #Tokens=160, #Dim.=128, Latency (s)=0.202026.0638.4929.050.07190.8831—
OmniTokenizerTokens (T×H×W)=5 × 16 × 16, Ratio=82026.0142————
OmniTokenizerTokenizer #Params=82.2M, Generator #Params=650M2026.0342———1,280
OmniTokenizer-CVToken type=Continuous, Total # tokens=8192, # Frames=172024.1149.528.30.0810.913—
LARP#Tokens=1024, #Dim.=16, T/P (%)=0.0982026.0651.4525.530.09730.8262—
CV-VAE#Tokens=4096, #Dim.=4, T/P (%)=0.3912026.0652.4329.470.06850.8849—
MAGVITTokens (T×H×W)=5 × 16 × 16, Ratio=82026.0158————
VideoAR-LTokens (T×H×W)=5 × 8 × 8, Ratio=162026.0161————
TivTok-T1024Video resolution=32×256×256, #Tokens=1280, #Dim.=162026.0661.4629.130.07110.8857—
OmniTokenizer#Tokens=1280, Codebook Size=8192, Resolution=128x128, Frames=162026.016227.760.112——
CV-VAEVideo resolution=32×256×256, #Tokens=8192, #Dim.=4, Latency (s)=1.782026.0664.2129.120.06920.8809—
PVDM-AEToken type=Continuous, Total # tokens=6144, # Frames=162024.1166.526.50.120.859—
OpenSora#Tokens=4096, #Dim.=16, T/P (%)=0.3912026.0667.52————
CV-VAEVideo resolution=128×256×256, #Tokens=32768, #Dim.=4, Latency (s)=7.122026.0672.91290.07290.8831—
CosmosTokenizer-CV*Token type=Continuous, Total # tokens=8192, # Frames=172024.1187.828.50.1190.905—
TivTok-T128Video resolution=128×256×256, #Tokens=352, #Dim.=128, Latency (s)=0.712026.0692.0926.230.10570.821—
OmniTokenizer-DVToken type=Discrete, Total # tokens=8192, # Frames=172024.1197.926.10.1130.871—
CoordTokToken type=Continuous, Total # tokens=1280, # Frames=1282024.11102.928.60.0660.892—
Cosmos-S#Tokens=512, #Dim.=16, T/P (%)=0.0492026.06104.5128.260.10460.8577—
Cosmos-Tokenizer-DV#Tokens=1280, Codebook Size=64000, Resolution=128x128, Frames=162026.01140————
TATS#Tokens=1024, Codebook Size=16384, Resolution=128x128, Frames=162026.0115724.220.206——
TATSTokens (T×H×W)=4 × 16 × 16, Ratio=82026.01162————
TATSTokenizer #Params=32M, Generator #Params=321M2026.03162———1,024
LARPToken type=Discrete, Total # tokens=8192, # Frames=162024.11201.324.30.1420.806—
LARPVideo resolution=32×256×256, #Tokens=2048, #Dim.=16, Latency (s)=1.752026.06226.7923.150.17570.7479—
ElasticTok#Tokens=1022, Codebook Size=64000, Resolution=128x128, Frames=162026.01230————
TATS-AEToken type=Discrete, Total # tokens=8192, # Frames=162024.11249.423.20.2130.792—
VidTwin#Tokens=1008, #Dim.=4/8, T/P (%)=0.1262026.06388.8628.140.24140.8044—
ElasticTok#Tokens=1024, #Dim.=16, T/P (%)=0.3912026.06390————
OmniTokenizer-CVToken type=Continuous, Total # tokens=1024, # Frames=172024.11396.723.20.1750.744—
LARPToken type=Discrete, Total # tokens=1024, # Frames=162024.11443.5220.1810.766—
MaskGIT-AEToken type=Discrete, Total # tokens=8192, # Frames=12024.11447.121.40.1390.667—
OmniTokenizer-DVToken type=Discrete, Total # tokens=1024, # Frames=172024.1150922.20.2010.703—
CosmosTokenizer-CV*Token type=Continuous, Total # tokens=1024, # Frames=172024.11519.6240.220.774—
CoordTokVideo resolution=128×128×128, #Tokens=1280, #Dim.=82026.061,108.7627.250.23460.7503—
PVDM-AEToken type=Continuous, Total # tokens=1152, # Frames=162024.111,270.119.10.3330.563—
LARPVideo resolution=128×256×256, #Tokens=8192, #Dim.=16, Latency (s)=22.782026.063,223.5514.850.62510.2924—
MAGVIT-AE-LToken type=Discrete, Total # tokens=8192, # Frames=162024.11—21.80.1130.69—