Image Generation on GenEval
89Overall ScoreWan-Weaver
Evaluation Results
| Method | Links | |
|---|---|---|
| Wan-WeaverModel Category=Unified Models2026.03 | 89 | |
| Wan-Weaver2026.03 | 89 | |
| BAGEL-Uni-Edit (Ours)Training Stage=Stage 12026.05 | 89 | |
| BAGEL-Uni-Edit (Ours)Training Stage=Stage 22026.05 | 89 | |
| DuoGenModel Type=Interleaved Generation2026.01 | 88 | |
| Bagel# Tokens=5.1T, # Params.=7B / 14B, Architecture Category=Autoregressive meets Diffusion, Prompt Rewriting (flag)=true2025.09 | 88 | |
| BagelModel Category=Unified Models, LLM rewriter=true2026.03 | 88 | |
| Bagel2026.03 | 88 | |
| BAGEL-RecA2026.05 | 88 | |
| Uni-Edit ViT (und.) 224x224Resolution=224x2242026.05 | 88 | |
| Uni-Edit ViT (und.) 224x980Resolution=224x9802026.05 | 88 | |
| Qwen-ImageModel Type=Generation2026.01 | 87 | |
| BAGEL2026.05 | 87 | |
| Uni-Edit w/o VAE dropoutVAE dropout=false2026.05 | 87 | |
| UniPicModel Category=Unified Models2026.03 | 86 | |
| OmniGen2Model Category=Unified Models, LLM rewriter=true2026.03 | 86 | |
| BAGEL-AnyEdit2026.05 | 86 | |
| Uni-Edit w/o joint trainingjoint training=false2026.05 | 86 | |
| Uni-Edit ViT (und.) 378x980Resolution=378x9802026.05 | 86 | |
| GPT-4o-ImageModel Type=Commercial2026.01 | 84 | |
| Blip-3oModel Type=Unified Model2026.01 | 84 | |
| BLIP3-oModel Category=Unified Models, LLM rewriter=true2026.03 | 84 | |
| UniWorld-V1Model Category=Unified Models, LLM rewriter=true2026.03 | 84 | |
| UniWorld-V12026.03 | 84 | |
| Uni-CoTModel Type=Interleaved Generation, Interleaved generation to improve image quality=true2026.01 | 83 | |
| X-Omni# Tokens=~1T, # Params.=10B / 20B, Architecture Category=Autoregressive meets Diffusion, Prompt Rewriting (flag)=true2025.09 | 83 | |
| Uni-X# Tokens=240B, # Params.=3B / 4.5B, Architecture Category=Autoregressive w/o Semantic Encoder, Prompt Rewriting (flag)=true, Extended Image-Text Training (flag)=true2025.09 | 83 | |
| FLUX.1-devModel Type=Generation2026.01 | 82 | |
| BagelModel Type=Unified Model2026.01 | 82 | |
| BAGEL-7BStage=Uni. 1 Stage2025.12 | 82 | |
| Uni-X# Tokens=140B, # Params.=3B / 4.5B, Architecture Category=Autoregressive w/o Semantic Encoder, Prompt Rewriting (flag)=true2025.09 | 82 | |
| +Uni-Edit (Ours)Base Model=Janus-Pro2026.05 | 82 | |
| Bilp3o# Tokens=-, # Params.=4B / 9B, Architecture Category=Autoregressive meets Diffusion2025.09 | 81 | |
| Janus-Pro-7BModel Type=Unified Model2026.01 | 80 | |
| MetaQuery-XLModel Type=Unified Model, LLM prompt rewriting=true2026.01 | 80 | |
| UniWorld-V1Model Type=Unified Model2026.01 | 80 | |
| OmniGen2Model Type=Unified Model2026.01 | 80 | |
| Janus-Pro-7BStage=Uni. 1 Stage2025.12 | 80 | |
| Janus-Pro# Tokens=~300B, # Params.=7B, Architecture Category=Autoregressive w/ Semantic Encoder2025.09 | 80 | |
| Janus-ProModel Category=Unified Models2026.03 | 80 | |
| MetaQuery-XLModel Category=Unified Models, LLM rewriter=true2026.03 | 80 | |
| Janus-Pro2026.05 | 80 | |
| MetaQuery-LStage=Uni. 2 Stage2025.12 | 78 | |
| Show-o2Model Category=Unified Models2026.03 | 76 | |
| Janus-Pro-1BStage=Uni. 1 Stage2025.12 | 73 | |
| NextStep1# Tokens=~1T, # Params.=14B, Architecture Category=Autoregressive w/ Semantic Encoder, Prompt Rewriting (flag)=true2025.09 | 73 | |
| UnisonStage=Uni. 2 Stage2025.12 | 70 | |
| Show-o# Tokens=~500B, # Params.=1.3B, Architecture Category=Autoregressive meets Diffusion, Semantic Alignment (flag)=true2025.09 | 69 | |
| Show-o# Tokens=~500B, # Params.=1.3B, Architecture Category=Autoregressive meets Diffusion2025.09 | 68 | |
| Liquid# Tokens=~90B, # Params.=7B, Architecture Category=Autoregressive w/o Semantic Encoder, Prompt Rewriting (flag)=true2025.09 | 68 | |
| OmniGenModel Category=Unified Models2026.03 | 68 | |
| DALLE-3Model Type=Generation2026.01 | 67 | |
| EMU3# Tokens=-, # Params.=8B, Architecture Category=Autoregressive w/o Semantic Encoder, Prompt Rewriting (flag)=true2025.09 | 66 | |
| FLUX.1-devModel Category=Text-to-Image Generation Models2026.03 | 66 | |
| Emu3Model Category=Unified Models, LLM rewriter=true2026.03 | 66 | |
| FLUX.1-dev2026.03 | 66 | |
| MMaDAModel Type=Unified Model2026.01 | 63 | |
| SD3-mediumModel Category=Text-to-Image Generation Models2026.03 | 62 | |
| SDXLModel Type=Generation2026.01 | 55 | |
| TokenFlow-13BStage=Uni. 1 Stage2025.12 | 55 | |
| SDXLModel Category=Text-to-Image Generation Models2026.03 | 55 | |
| Emu3Model Type=Unified Model2026.01 | 54 | |
| Show-oModel Type=Unified Model2026.01 | 53 | |
| Show-oStage=Uni. 1 Stage2025.12 | 53 | |
| HART#Scales=14, FlashAttention=true, Resolution=1024x10242026.02 | 50.9 | |
| SparVAR#Scales=14, Resolution=1024x10242026.02 | 50.7 | |
| FastVAR#Scales=14, Resolution=1024x10242026.02 | 50.4 | |
| LWM# Tokens=~500B, # Params.=7B, Architecture Category=Autoregressive w/o Semantic Encoder2025.09 | 47 | |
| Chameleon# Tokens=9.2T, # Params.=34B, Architecture Category=Autoregressive w/o Semantic Encoder2025.09 | 39 |