Visual Text Generation on SimpleBench
42AccuracyGlyphControl
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| GlyphControl#Params=1.3B, Text Encoder=CLIP(354M), Training Dataset=LAION-Glyph-10M2023.05 | 42 | 48 | 1.43 | |
| GlyphControl#Params=1.3B, Text Encoder=CLIP(354M), Training Dataset=LAION-Glyph-1M2023.05 | 40 | 45 | 1.59 | |
| GlyphControl#Params=1.3B, Text Encoder=CLIP(354M), Training Dataset=LAION-Glyph-100K2023.05 | 30 | 37 | 1.77 | |
| DeepFloyd (IF-I-XL)#Params=6.0B, Text Encoder=T5-XXL(4.8B), Training Dataset=LAION 1.2B2023.05 | 0.6 | 33 | 1.63 | |
| SDXL 1.0#Params=5.8B, Text Encoder=CLIP & OpenCLIP(817M), Training Dataset=Internal Dataset (>100M)2023.05 | 0.3 | 13 | 6.26 | |
| DeepFloyd (IF-I-M)#Params=2.1B, Text Encoder=T5-XXL(4.8B), Training Dataset=LAION 1.2B2023.05 | 0.3 | 18 | 2.44 | |
| DeepFloyd (IF-I-L)#Params=2.6B, Text Encoder=T5-XXL(4.8B), Training Dataset=LAION 1.2B2023.05 | 0.3 | 26 | 1.97 | |
| Stable Diffusion v2.0#Params=865M, Text Encoder=CLIP(354M), Training Dataset=LAION 1.2B2023.05 | 0 | 3 | 4.25 |