List your company
ResearchFormerly WizWand

Image Generation on DPG (DPG Metric)

88.32DPG ScoreQwen-Image
66.42872.111577.79583.4785Sep 29, 2025Oct 28, 2025Nov 27, 2025Dec 27, 2025Jan 25, 2026Feb 24, 2026Mar 26, 2026
Updated 7mo ago

Evaluation results

MethodLinks
Qwen-ImageParams=7B+20B, Types=AR+Diff, Architecture Category=Gen. Only2025.1188.32
X-Omni# Tokens=~1T, # Params.=10B / 20B, Architecture Category=Autoregressive meets Diffusion2025.0987.6
Wan-WeaverModel Category=Unified Models2026.0387.21
Wan-Weaver2026.0387.21
InfinityParams=8B, Types=VAR, Architecture Category=Gen. Only2025.1186.6
Show-o2Params=7B, Types=AR+Diff, Architecture Category=Uni. Double Image Encoders / Training Objectives2025.1186.14
Show-o2Model Category=Unified Models2026.0386.14
UniPicModel Category=Unified Models2026.0385.5
NextStep1# Tokens=~1T, # Params.=14B, Architecture Category=Autoregressive w/ Semantic Encoder2025.0985.2
BAGELParams=7B+7B, Types=AR+Diff, Architecture Category=Uni. MoE/MoT Arch2025.1185.07
BagelModel Category=Unified Models2026.0385.07
Bagel2026.0385.07
Bagel# Tokens=5.1T, # Params.=7B / 14B, Architecture Category=Autoregressive meets Diffusion2025.0985
OneCatParams=3B+3B+3B, Types=AR+VAR, Architecture Category=Uni. MoE/MoT Arch2025.1184.53
Janus-Pro + AIA (Ours)Params=7B, Types=AR, Architecture Category=Uni. Double Image Encoders / Training Objectives2025.1184.49
Janus-ProParams=7B, Types=AR, Architecture Category=Uni. Double Image Encoders / Training Objectives2025.1184.19
Janus-ProModel Category=Unified Models2026.0384.19
Janus-Pro# Tokens=~300B, # Params.=7B, Architecture Category=Autoregressive w/ Semantic Encoder2025.0984.1
SD3-mediumModel Category=Text-to-Image Generation Models2026.0384.1
SD3-mediumParams=2B, Types=Diff, Architecture Category=Gen. Only2025.1184.08
FLUX.1-devParams=12B, Types=Diff, Architecture Category=Gen. Only2025.1184
FLUX.1-devModel Category=Text-to-Image Generation Models2026.0384
FLUX.1-dev2026.0384
OmniGen2Params=3B+4B, Types=AR+Diff, Architecture Category=Uni. Frozen MLLM2025.1183.57
OmniGen2Model Category=Unified Models2026.0383.57
InfinityParams=2B, Types=VAR, Architecture Category=Gen. Only2025.1183.5
MetaQuery-XLParams=7B+1.6B, Types=AR+Diff, Architecture Category=Uni. Frozen MLLM2025.1182.05
MetaQuery-XLModel Category=Unified Models2026.0382.05
Blip3-oParams=7B+1.4B, Types=AR+Diff, Architecture Category=Uni. Frozen MLLM2025.1181.6
BLIP3-oModel Category=Unified Models2026.0381.6
UniWorld-V1Params=7B+12B, Types=AR+Diff, Architecture Category=Uni. Frozen MLLM2025.1181.38
UniWorld-V1Model Category=Unified Models2026.0381.38
UniWorld-V12026.0381.38
Emu3 + AIA (Ours)Params=8B, Types=AR, Architecture Category=Uni. Purely2025.1181.2
OmniGenModel Category=Unified Models2026.0381.16
Emu3-GenParams=8B, Types=AR, Architecture Category=Gen. Only2025.1180.6
EMU3# Tokens=-, # Params.=8B, Architecture Category=Autoregressive w/o Semantic Encoder2025.0980.6
Emu3Model Category=Unified Models2026.0380.6
Uni-X# Tokens=240B, # Params.=3B / 4.5B, Architecture Category=Autoregressive w/o Semantic Encoder, Extended Image-Text Training (flag)=true2025.0980.3
Liquid# Tokens=~90B, # Params.=7B, Architecture Category=Autoregressive w/o Semantic Encoder2025.0979.8
Uni-X# Tokens=140B, # Params.=3B / 4.5B, Architecture Category=Autoregressive w/o Semantic Encoder2025.0979.8
Bilp3o# Tokens=-, # Params.=4B / 9B, Architecture Category=Autoregressive meets Diffusion2025.0979.3
Emu3Params=8B, Types=AR, Architecture Category=Uni. Purely, re-implemented=true2025.1179.24
SDXLModel Category=Text-to-Image Generation Models2026.0374.7
SDXLParams=-, Types=Diff, Architecture Category=Gen. Only2025.1174.65
VILA-UParams=7B, Types=AR, Architecture Category=Uni. Purely2025.1172.48
Show-oParams=1.3B, Types=AR+Mask, Architecture Category=Uni. Double Image Encoders / Training Objectives2025.1167.27