Hugging Face's logo

bghira
/
pseudo-flex-base

Text-to-Image
Diffusers
Safetensors
StableDiffusionPipeline
stable-diffusion
stable-diffusion-2-1
Model card Files Community
PseudoTerminal X commited on
Commit
86388b0
·
1 Parent(s): edbab9f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -6
README.md CHANGED
@@ -14,13 +14,12 @@ library_name: diffusers
14
  stable-diffusion-2-1 (stabilityai/stable-diffusion-2-1) finetuned with different aspect ratios, into a photography model (ptx0/pseudo-real-beta).
15
 
16
 
17
- <!--
18
- | resolution | model | stable diffusion | pseudo-flex | realism-engine |
19
- |:---------------:|:-------:|:----------------------------:|:-----------------------------:|:-------------------------------:
20
- | 576x1024 (9:16) | v2-1 | ![img](imgs/21-576-1024.png) | ![img](imgs/21f-576-1024.png) | ![img](imgs/ellen-flex.png) |
21
- | 1280x720 (16:9) | v2-1 | ![img](imgs/ellen-base.png) | ![img](imgs/ellen-flex.png) | ![img](imgs/ellen-realism.png) |
22
 
23
- -->
24
  ## Background
25
 
26
  The `ptx0/pseudo-real-beta` pretrained checkpoint had its unet trained for 4,200 steps and its text encoder trained for 15,600 steps at a batch size of 15 with 10 gradient accumulations, on a diverse dataset:
 
14
  stable-diffusion-2-1 (stabilityai/stable-diffusion-2-1) finetuned with different aspect ratios, into a photography model (ptx0/pseudo-real-beta).
15
 
16
 
17
+ | resolution | model | stable diffusion | pseudo-flex | realism-engine |
18
+ |:---------------:|:-------:|:------------------------------:|:-------------------------------:|:---------------------------------:
19
+ | 576x1024 (9:16) | v2-1 | ![img](assets/21-576-1024.png) | ![img](assets/21f-576-1024.png) | ![img](assets/ellen-flex.png) |
20
+ | 1280x720 (16:9) | v2-1 | ![img](assets/ellen-base.png) | ![img](assets/ellen-flex.png) | ![img](assets/ellen-realism.png) |
21
+
22
 
 
23
  ## Background
24
 
25
  The `ptx0/pseudo-real-beta` pretrained checkpoint had its unet trained for 4,200 steps and its text encoder trained for 15,600 steps at a batch size of 15 with 10 gradient accumulations, on a diverse dataset: