Text-to-Image Generation on LAION-5B 1,000 prompts benchmark leaderboard · SOTA2 Research

Text-to-Image Generation on LAION-5B 1,000 prompts

47.079FID (Real)DemoFusion

Evaluation Results

Method	Links
DemoFusion2025.03		47.079	0.002	19.533	26.441	0.004	27.843	33.748	79	—
AccDiffusion2025.03		48.143	0.002	18.466	32.747	0.008	24.778	33.153	111	13.8
RectifiedHR2025.03		48.361	0.002	20.616	25.347	0.003	28.126	33.756	13	32.2
RectifiedHR2025.03		48.684	0.003	20.352	35.718	0.009	20.819	33.415	37	32.2
DiffuseHigh2025.03		48.861	0.003	19.716	40.267	0.01	21.55	33.39	190	—
DemoFusion2025.03		48.983	0.003	18.225	38.136	0.01	20.786	33.311	605	—
FreCas2025.03		49.129	0.003	20.274	27.002	0.004	29.843	33.7	14	16.2
DiffuseHigh2025.03		49.748	0.003	19.537	27.667	0.004	27.876	33.436	37	—
FreCas2025.03		49.764	0.003	18.656	39.047	0.01	21.7	33.237	74	16.2
AccDiffusion2025.03		54.918	0.005	17.444	60.362	0.023	16.37	32.438	826	13.8
ElasticDiffusion2025.03		56.639	0.01	15.326	37.649	0.014	19.867	32.301	150	—
CutDiffusion2025.03		59.152	0.007	17.109	38.004	0.008	23.444	32.573	53	—
HiDiffusion2025.03		63.674	0.007	16.876	41.93	0.008	23.165	31.711	18	12.7
ScaleCrafter2025.03		64.236	0.007	15.952	45.861	0.01	22.252	31.803	35	13.6
FouriScale2025.03		71.344	0.01	15.957	53.99	0.014	20.625	31.157	59	11.6
HiDiffusion2025.03		93.515	0.024	11.878	120.17	0.058	11.272	27.853	71	12.7
ElasticDiffusion2025.03		101.313	0.056	9.406	111.102	0.089	7.627	27.725	400	—
ScaleCrafter2025.03		110.094	0.028	10.098	112.105	0.043	11.421	27.809	528	13.6
CutDiffusion2025.03		130.207	0.055	9.334	113.033	0.055	10.961	26.734	193	—
FouriScale2025.03		135.111	0.046	9.481	129.895	0.057	9.792	26.891	489	11.6