List your company

Text-to-image generation on MS-COCO 2014 (val)

2.47FIDRandom Train Images
1.44888.341915.23522.1281Jun 22, 2022Jan 22, 2023Aug 25, 2023Mar 26, 2024Oct 27, 2024May 29, 2025Dec 30, 2025

Evaluation results

MethodLinks
Random Train ImagesEvaluation Protocol=Zero-shot, Guidance Scale=1.22022.062.47—————————————
PartiModel Type=Autoregressive, Evaluation Protocol=Finetuned, Guidance Scale=1.22022.063.22—————————————
RAPHAELData distribution=English, Resolution=256 × 256, Evaluation protocol=zero-shot, Sample size=30k2023.096.61—————————————
ERNIE-ViLG 2.0Architecture=Non-AR based, Parameters=24B2023.036.75—————————————
ERNIE-ViLG 2.0Data distribution=English and Chinese, Resolution=256 × 256, Evaluation protocol=zero-shot, Sample size=30k2023.096.75—————————————
Retrieval BaselineEvaluation Protocol=Finetuned, Guidance Scale=1.22022.066.82—————————————
Re-Imagenzero-shot=true, num_samples=300002023.056.88—————————————
eDiff-IData distribution=English, Resolution=256 × 256, Evaluation protocol=zero-shot, Sample size=30k2023.096.95—————————————
PartiModel Type=Autoregressive, Evaluation Protocol=Zero-shot, Guidance Scale=1.22022.067.23—————————————
Partizero-shot=true, num_samples=300002023.057.23—————————————
ImagenModel Type=Diffusion, Evaluation Protocol=Zero-shot, Guidance Scale=1.22022.067.27—————————————
ImagenArchitecture=Non-AR based, Parameters=4.9B2023.037.27—————————————
Imagenzero-shot=true, num_samples=300002023.057.27—————————————
Make-A-SceneModel Type=Autoregressive, Evaluation Protocol=Finetuned, Guidance Scale=1.22022.067.55—————————————
LAFITE2zero-shot=true, num_samples=300002023.058.42—————————————
Stable DiffusionData distribution=English, Resolution=256 × 256, Evaluation protocol=zero-shot, Sample size=30k2023.098.59—————————————
SD v1.5zero-shot=true, num_samples=30000, classifier-free guidance scaling factor=3, DDIM inference steps=2502023.059.22—————————————
XMC-GANModel Type=GAN, Evaluation Protocol=Finetuned, Guidance Scale=1.22022.069.33—————————————
BDMData distribution=Chinese, Resolution=256 × 256, Evaluation protocol=zero-shot, Sample size=30k, classifier-free guidance scale=3.7, Chinese language branch weight=1.5, backbone=SD1.52023.099.93—————————————
DALL-E 2Model Type=Diffusion, Evaluation Protocol=Zero-shot, Guidance Scale=1.22022.0610.39—————————————
DALL-E-2Architecture=Non-AR based, Parameters=6.5B2023.0310.39—————————————
DALL-E 2zero-shot=true, num_samples=300002023.0510.39—————————————
DALL-E 2Data distribution=English, Resolution=256 × 256, Evaluation protocol=zero-shot, Sample size=30k2023.0910.39—————————————
MAGVLT (T2I only)Architecture=Non-AR based, Parameters=447M, Modality=T2I only2023.0310.74——23.948.12—————————
DALL-E 2zero-shot=true2023.0410.9————10.53871——————
Make-A-SceneModel Type=Autoregressive, Evaluation Protocol=Zero-shot, Guidance Scale=1.22022.0611.84—————————————
Make-A-SceneArchitecture=AR based, Parameters=4B2023.0311.84—————————————
Make-A-Scenezero-shot=true, num_samples=300002023.0511.84—————————————
MAGVLTParameters=447M, Modality=Available for both T2I & I2T2023.0312.08——22.758.12—————————
GILLzero-shot=true, num_samples=30000, classifier-free guidance scaling factor=3, DDIM inference steps=2502023.0512.2—————————————
GLIDEModel Type=Diffusion, Evaluation Protocol=Zero-shot, Guidance Scale=1.22022.0612.24—————————————
GLIDEArchitecture=Non-AR based, Parameters=3.5B2023.0312.24—————————————
GLIDEzero-shot=true, num_samples=300002023.0512.24—————————————
GLIDEData distribution=English, Resolution=256 × 256, Evaluation protocol=zero-shot, Sample size=30k2023.0912.24—————————————
LDMData distribution=English, Resolution=256 × 256, Evaluation protocol=zero-shot, Sample size=30k2023.0912.61—————————————
Parti-350MArchitecture=AR based, Parameters=350M2023.0314.1—————————————
ERNIE-ViLGData distribution=English and Chinese, Resolution=256 × 256, Evaluation protocol=zero-shot, Sample size=30k2023.0914.7—————————————
DPM-Solver-v3NFE=12, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.13—0.265———————————
DPM-Solver-v3NFE=20, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.23—0.265———————————
DPM-Solver-v3NFE=15, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.3—0.265———————————
DPM-Solver-v3NFE=10, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.32—0.265———————————
DPM-Solver-v3NFE=8, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.41—0.264———————————
DPM-Solver++NFE=20, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.72—0.265———————————
DPM-Solver++NFE=12, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.78—0.266———————————
DPM-Solver++NFE=15, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.84—0.265———————————
DPM-Solver++NFE=10, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.93—0.265———————————
UniPCNFE=20, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1015.94—0.265———————————
UniPCNFE=15, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1016.06—0.265———————————
UniPCNFE=12, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1016.09—0.265———————————
UniPCNFE=10, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1016.15—0.265———————————
UniPCNFE=8, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1016.2—0.265———————————
DPM-Solver++NFE=8, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1016.4—0.265———————————
DPM-Solver-v3NFE=6, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1016.41—0.262———————————
ARGVLTParameters=447M, Modality=Available for both T2I & I2T2023.0316.93——22.51—————————
UniPCNFE=6, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1017.32—0.263———————————
DPM-Solver++NFE=6, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1017.44—0.265———————————
CogView2Model Type=Autoregressive, Evaluation Protocol=Finetuned, Guidance Scale=1.22022.0617.7—————————————
Stable Diffusion 1.4Resolution=256x256, zero-shot=true2023.0417.9————4-0.08572——————
Retrieval BaselineEvaluation Protocol=Zero-shot, Guidance Scale=1.22022.0617.97—————————————
Versatile DiffusionResolution=256x256, zero-shot=true2023.0418.4————5-0.54853——————
UniPCNFE=5, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1018.77—0.262———————————
Stable Diffusion 2.1-baseResolution=256x256, zero-shot=true2023.0418.8————20.15534——————
DPM-Solver-v3NFE=5, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1018.83—0.26———————————
DPM-Solver++NFE=5, Guidance scale=7.5, Model=Stable-Diffusion [43]2023.1018.87—0.263———————————
DeepCacheN=22025.1219.44—————1.07—22.850.241.29———
∆-DiTb=302025.1219.72—————1.06—22.850.251.62———
Stable Diffusion 3.5 LargeSampling steps=502025.1220.01—————1.09—22.9—————
BlockDanceN=32025.1220.26—————1.08—22.890.041.49———
OpenjourneyResolution=256x256, zero-shot=true2023.0420.7————3-0.04555——————
CorGi+D=5, γ=30, δ=22025.1220.7—————1.05—22.760.082.02———
CorGi+D=6, γ=28, δ=22025.1220.71—————1.04—22.740.092.02———
CorGiD=5, γ=30, δ=22025.1221.38—————1.02—22.740.082.09———
CorGiD=6, γ=28, δ=22025.1221.45—————1.02—22.730.092.08———
RASratio=30%2025.1221.5—————1—22.450.111.99———
ARGVLT (T2I only)Architecture=AR based, Parameters=447M, Modality=T2I only2023.0321.8——19.271—————————
CogView2Model Type=Autoregressive, Evaluation Protocol=Zero-shot, Guidance Scale=1.22022.0624—————————————
CogView2Architecture=AR based, Parameters=6B2023.0324——22.4——————————
SD-1.5Resolution=512x5122025.0924.01—30.17———————————
SD-2.1-baseResolution=512x5122025.0924.25—31.32———————————
MaskAttn-SDXLSetting=mid-res, default, Resolution=1024x10242025.0924.57—31.75———————————
FireFlowSteps=10, NFE=11, ODE Solver=2nd-order2024.1225.16—31.42———————————
PixArt-ΣInference Setting=method-recommended2025.0925.23—31.55———————————
PixArt-αInference Setting=method-recommended2025.0925.68—31.49———————————
SDXLResolution=1024x10242025.0925.77—31.53———————————
RF-SolverSteps=10, NFE=20, ODE Solver=2nd-order2024.1225.93—31.35———————————
CogView 2Resolution=256x256, zero-shot=true2023.0426.2————6-0.8516——————
DeepCacheBlock-segment caching (N)=2, Speedup=1.26×2025.1226.24————1.09——23.110.25————
BlockDanceBlock-segment caching (N)=3, Speedup=1.50×2025.1226.24————1.1——23.120.05————
FLUX-devSteps=20, NFE=20, ODE Solver=1st-order2024.1226.77—31.44———————————
CogViewModel Type=Autoregressive, Evaluation Protocol=Zero-shot, Guidance Scale=1.22022.0627.1—————————————
CogViewArchitecture=AR based, Parameters=4B2023.0327.1——18.2——————————
OBS-DiffBase Model=SD 3-medium, Sparsity (%)=502025.1027.2—0.3167———0.6468———————
OBS-DiffBase Model=SD v2.1-base, Sparsity (%)=502025.1027.41—0.3102———-0.0356———————
CorGi+Caching parameters (D, γ, δ)=D=6, γ=47, δ=2, Speedup=1.98×2025.1227.48————1.01——22.820.12————
DALL-EArchitecture=AR based, Parameters=12B2023.0327.5——17.9——————————
CorGi+Caching parameters (D, γ, δ)=D=5, γ=49, δ=2, Speedup=1.91×2025.1227.56————1.05——22.950.09————
FLUX.1-devInference steps=302025.1227.77————1.08——23.11—————
MagnitudeBase Model=SD v2.1-base, Sparsity (%)=402025.1027.86—0.3111———0.1864———————
WandaBase Model=SD v2.1-base, Sparsity (%)=402025.1027.96—0.3122———0.1367———————
DALL-EModel Type=Autoregressive, Evaluation Protocol=Zero-shot, Guidance Scale=1.22022.0628—————————————