Update README.md
Browse files
README.md
CHANGED
|
@@ -8,19 +8,21 @@ tags:
|
|
| 8 |
---
|
| 9 |
## Model Card for lyraSD
|
| 10 |
|
| 11 |
-
|
|
|
|
|
|
|
| 12 |
|
| 13 |
Among its main features are:
|
| 14 |
|
|
|
|
|
|
|
|
|
|
| 15 |
- 4 Commonly used Pipelines
|
| 16 |
- - Text2Img
|
| 17 |
- - Img2Img
|
| 18 |
- - ControlNetText2Img
|
| 19 |
- - ControlNetImg2Img
|
| 20 |
-
-
|
| 21 |
-
- ControlNet Hot Swap: Can hot swap a ControlNet model weights within 0.4s (0s if cached)
|
| 22 |
-
- Lora How Swap: Can hot swap a Lora within 0.5s (0.1s if cached)
|
| 23 |
-
- device requirements: Nvidia Ampere architecture (A100, A10) or compatable
|
| 24 |
|
| 25 |
## Speed
|
| 26 |
|
|
|
|
| 8 |
---
|
| 9 |
## Model Card for lyraSD
|
| 10 |
|
| 11 |
+
We consider the Diffusers as the much more extendable framework for the SD ecosystem. Therefore, we have made a pivot to Diffusers, leading to a complete update of lyraSD.
|
| 12 |
+
|
| 13 |
+
lyraSD is currently the **fastest Stable Diffusion model** that can 100% align the outputs of **Diffusers** available, boasting an inference cost of only **0.52 seconds** for a 512x512 image, accelerating the process up to **80% faster** than the original version.
|
| 14 |
|
| 15 |
Among its main features are:
|
| 16 |
|
| 17 |
+
- **ControlNet Hot Swap**: Can hot swap a ControlNet model weights within 0.4s (0s if cached)
|
| 18 |
+
- **Lora How Swap**: Can hot swap a Lora within 0.5s (0.1s if cached)
|
| 19 |
+
- 100% likeness to diffusers output
|
| 20 |
- 4 Commonly used Pipelines
|
| 21 |
- - Text2Img
|
| 22 |
- - Img2Img
|
| 23 |
- - ControlNetText2Img
|
| 24 |
- - ControlNetImg2Img
|
| 25 |
+
- Device requirements: Nvidia Ampere architecture (A100, A10) or compatable
|
|
|
|
|
|
|
|
|
|
| 26 |
|
| 27 |
## Speed
|
| 28 |
|