Update README.md
Browse files
README.md
CHANGED
|
@@ -21,6 +21,16 @@ tags:
|
|
| 21 |
# KOALA-1B Model Card
|
| 22 |
|
| 23 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
## Abstract
|
| 25 |
### TL;DR
|
| 26 |
> We propose a fast text-to-image model, called KOALA, by compressing SDXL's U-Net and distilling knowledge from SDXL into our model. KOALA-700M can generate a 1024x1024 image in less than 1.5 seconds on an NVIDIA 4090 GPU, which is more than 2x faster than SDXL. KOALA-700M can be used as a decent alternative between SDM and SDXL in limited resources.
|
|
|
|
| 21 |
# KOALA-1B Model Card
|
| 22 |
|
| 23 |
|
| 24 |
+
## KOALA Model Cards
|
| 25 |
+
|
| 26 |
+
|Model|link|
|
| 27 |
+
|:--|:--|
|
| 28 |
+
|koala-700m | https://huggingface.co/etri-vilab/koala-700m|
|
| 29 |
+
|koala-700m-llava-cap | https://huggingface.co/etri-vilab/koala-700m-llava-cap|
|
| 30 |
+
|koala-1b | https://huggingface.co/etri-vilab/koala-1bm|
|
| 31 |
+
|koala-1b-llava-cap | https://huggingface.co/etri-vilab/koala-1b-llava-cap|
|
| 32 |
+
|
| 33 |
+
|
| 34 |
## Abstract
|
| 35 |
### TL;DR
|
| 36 |
> We propose a fast text-to-image model, called KOALA, by compressing SDXL's U-Net and distilling knowledge from SDXL into our model. KOALA-700M can generate a 1024x1024 image in less than 1.5 seconds on an NVIDIA 4090 GPU, which is more than 2x faster than SDXL. KOALA-700M can be used as a decent alternative between SDM and SDXL in limited resources.
|