Text-to-Image Alignment on MS COCO
31.58CLIP ScoreAC
Evaluation Results
| Method | Links | |
|---|---|---|
| AC2026.01 | 31.58 | |
| SD v1.52026.01 | 31.42 | |
| MACE2026.01 | 31.05 | |
| SPM2026.01 | 30.77 | |
| FMN2026.01 | 30.56 | |
| ESD2026.01 | 30.12 | |
| SalUn2026.01 | 29.93 | |
| FIA2026.01 | 29.56 | |
| UCE2026.01 | 29.25 | |
| CP2026.01 | 27.93 | |
| D2POSteps=7, Solver=iPNDM2026.07 | 26.57 | |
| D2POSteps=6, Solver=iPNDM2026.07 | 26.52 | |
| LD3Steps=7, Solver=iPNDM2026.07 | 26.37 | |
| D2POSteps=5, Solver=iPNDM2026.07 | 26.3 | |
| LD3Steps=6, Solver=iPNDM2026.07 | 26.22 | |
| LD3Steps=5, Solver=iPNDM2026.07 | 26.21 | |
| D2POSteps=4, Solver=iPNDM2026.07 | 26.01 | |
| LD3Steps=4, Solver=iPNDM2026.07 | 25.87 | |
| hub text identificationModel=apple/DFN5B-CLIP-ViT-H-14-378, Text Category=Hub text2026.04 | 1.023 | |
| GLSModel=apple/DFN5B-CLIP-ViT-H-14-378, Text Category=Hub text2026.04 | 0.995 | |
| hub text identificationModel=apple/DFN5B-CLIP-ViT-H-14, Text Category=Hub text2026.04 | 0.974 | |
| GLSModel=apple/DFN5B-CLIP-ViT-H-14, Text Category=Hub text2026.04 | 0.965 | |
| hub text identificationModel=openai/clip-vit-base-patch32, Text Category=Hub text2026.04 | 0.842 | |
| HumanModel=apple/DFN5B-CLIP-ViT-H-14, Text Category=Caption text2026.04 | 0.838 | |
| HumanModel=apple/DFN5B-CLIP-ViT-H-14-378, Text Category=Caption text2026.04 | 0.837 | |
| hub text identificationModel=laion/CLIP-ViT-H-14-laion2B-s32B-b79K, Text Category=Hub text2026.04 | 0.825 | |
| hub text identificationModel=apple/DFN2B-CLIP-ViT-L-14, Text Category=Hub text2026.04 | 0.814 | |
| BLIP-2Model=apple/DFN5B-CLIP-ViT-H-14, Text Category=Caption text2026.04 | 0.789 | |
| hub text identificationModel=laion/CLIP-ViT-L-14-laion2B-s32B-b82K, Text Category=Hub text2026.04 | 0.782 | |
| GLSModel=laion/CLIP-ViT-H-14-laion2B-s32B-b79K, Text Category=Hub text2026.04 | 0.78 | |
| BLIP-2Model=apple/DFN5B-CLIP-ViT-H-14-378, Text Category=Caption text2026.04 | 0.777 | |
| HumanModel=laion/CLIP-ViT-H-14-laion2B-s32B-b79K, Text Category=Caption text2026.04 | 0.771 | |
| GLSModel=laion/CLIP-ViT-g-14-laion2B-s12B-b42K, Text Category=Hub text2026.04 | 0.767 | |
| HumanModel=openai/clip-vit-base-patch32, Text Category=Caption text2026.04 | 0.759 | |
| HumanModel=laion/CLIP-ViT-L-14-laion2B-s32B-b82K, Text Category=Caption text2026.04 | 0.748 | |
| hub text identificationModel=laion/CLIP-ViT-g-14-laion2B-s12B-b42K, Text Category=Hub text2026.04 | 0.747 | |
| BLIP-2Model=openai/clip-vit-base-patch32, Text Category=Caption text2026.04 | 0.739 | |
| GLSModel=openai/clip-vit-base-patch32, Text Category=Hub text2026.04 | 0.732 | |
| BLIP-2Model=laion/CLIP-ViT-H-14-laion2B-s32B-b79K, Text Category=Caption text2026.04 | 0.729 | |
| GLSModel=apple/DFN2B-CLIP-ViT-L-14, Text Category=Hub text2026.04 | 0.723 | |
| HumanModel=apple/DFN2B-CLIP-ViT-L-14, Text Category=Caption text2026.04 | 0.722 | |
| HumanModel=laion/CLIP-ViT-g-14-laion2B-s12B-b42K, Text Category=Caption text2026.04 | 0.721 | |
| BLIP-2Model=laion/CLIP-ViT-L-14-laion2B-s32B-b82K, Text Category=Caption text2026.04 | 0.72 | |
| hub text identificationModel=openai/clip-vit-large-patch14-336, Text Category=Hub text2026.04 | 0.701 | |
| GLSModel=laion/CLIP-ViT-L-14-laion2B-s32B-b82K, Text Category=Hub text2026.04 | 0.69 | |
| BLIP-2Model=laion/CLIP-ViT-g-14-laion2B-s12B-b42K, Text Category=Caption text2026.04 | 0.688 | |
| BLIP-2Model=apple/DFN2B-CLIP-ViT-L-14, Text Category=Caption text2026.04 | 0.678 | |
| GLSModel=openai/clip-vit-large-patch14-336, Text Category=Hub text2026.04 | 0.677 | |
| HumanModel=openai/clip-vit-large-patch14-336, Text Category=Caption text2026.04 | 0.654 | |
| hub text identificationModel=openai/clip-vit-large-patch14, Text Category=Hub text2026.04 | 0.649 | |
| hub text identificationModel=BAAI/AltCLIP, Text Category=Hub text2026.04 | 0.643 | |
| HumanModel=openai/clip-vit-large-patch14, Text Category=Caption text2026.04 | 0.639 | |
| HumanModel=BAAI/AltCLIP, Text Category=Caption text2026.04 | 0.635 | |
| GLSModel=openai/clip-vit-large-patch14, Text Category=Hub text2026.04 | 0.633 | |
| BLIP-2Model=openai/clip-vit-large-patch14-336, Text Category=Caption text2026.04 | 0.628 | |
| BLIP-2Model=BAAI/AltCLIP, Text Category=Caption text2026.04 | 0.622 | |
| BLIP-2Model=openai/clip-vit-large-patch14, Text Category=Caption text2026.04 | 0.613 | |
| GLSModel=BAAI/AltCLIP, Text Category=Hub text2026.04 | 0.512 | |
| BaselineModel Version=SD v1.4, CLIP Backbone=ViT-B/322026.03 | 0.32 | |
| AFM-curveModel Version=SD v1.5, CLIP Backbone=ViT-B/322026.03 | 0.319 | |
| BaselineModel Version=SD v1.5, CLIP Backbone=ViT-B/322026.03 | 0.318 | |
| AFM-curveModel Version=SD v1.4, CLIP Backbone=ViT-B/322026.03 | 0.318 | |
| AFM-curve + entropyModel Version=SD v1.4, CLIP Backbone=ViT-B/322026.03 | 0.317 | |
| AFM-curve + entropyModel Version=SD v1.5, CLIP Backbone=ViT-B/322026.03 | 0.316 | |
| FreeUModel Version=SD v1.5, CLIP Backbone=ViT-B/322026.03 | 0.307 | |
| FreeUModel Version=SD v1.4, CLIP Backbone=ViT-B/322026.03 | 0.307 | |
| SAGModel Version=SD v1.5, CLIP Backbone=ViT-B/322026.03 | 0.305 | |
| SAGModel Version=SD v1.4, CLIP Backbone=ViT-B/322026.03 | 0.304 |