Hugging Face's logo

segmind
/
Segmind-Vega

Text-to-Image
Diffusers
Safetensors
StableDiffusionXLPipeline
ultra-realistic
stable-diffusion
distilled-model
knowledge-distillation
Model card Files
xet
 Community
4
harishp commited on
Commit
b8f09ec
·
1 Parent(s): 735c404

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -6
README.md CHANGED
@@ -35,6 +35,14 @@ Employing a knowledge distillation strategy, Segmind-Vega leverages the teaching
35
 
36
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/62039c2d91d53938a643317d/bZylkXH3PhFhLYJWG6WJ5.png)
37
 
 
 
 
 
 
 
 
 
38
  ## Usage:
39
  This model can be used via the 🧨 Diffusers library.
40
 
@@ -98,12 +106,6 @@ These are the key hyperparameters used during training:
98
  - Image resolution: 1024
99
  - Mixed-precision: fp16
100
 
101
- ### Speed Comparison
102
-
103
- Segmind-Vega has demonstrated an impressive 100% speedup compared to the Base SDXL Model. Below is a comparison on an A100 80GB.
104
-
105
- Below are the speed-up metrics on an RTX 4090 GPU.
106
-
107
 
108
  ### Model Sources
109
 
 
35
 
36
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/62039c2d91d53938a643317d/bZylkXH3PhFhLYJWG6WJ5.png)
37
 
38
+ ## Speed Comparison (Segmind-Vega vs SD-1.5 vs SDXL)
39
+
40
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/62039c2d91d53938a643317d/9ec2r_wzlHtAjVhbm1Tpf.png)
41
+
42
+ ## Parameters Comparison (Segmind-Vega vs SD-1.5 vs SDXL)
43
+
44
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/62039c2d91d53938a643317d/vjS5mxFuB1PSS7TE5RbUm.png)
45
+
46
  ## Usage:
47
  This model can be used via the 🧨 Diffusers library.
48
 
 
106
  - Image resolution: 1024
107
  - Mixed-precision: fp16
108
 
 
 
 
 
 
 
109
 
110
  ### Model Sources
111