Update README.md
Browse files
README.md
CHANGED
|
@@ -70,6 +70,10 @@ The main issues in this case are conditioning for Phantom adds a latent with eac
|
|
| 70 |
|
| 71 |
Merging the models themselves a simple merge is what I used for the majority of my testing adding only the VACE blocks - but you can add small amounts to various Phantom layers to get improvement. Just be careful.
|
| 72 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
Ill try to update this further as things come to mind.
|
| 74 |
|
| 75 |
|
|
|
|
| 70 |
|
| 71 |
Merging the models themselves a simple merge is what I used for the majority of my testing adding only the VACE blocks - but you can add small amounts to various Phantom layers to get improvement. Just be careful.
|
| 72 |
|
| 73 |
+
|
| 74 |
+
|
| 75 |
+
In terms of speedup the Lightx LORAs cause the model to follow the prompt more I think but this ends up breaking the conditioning and often results in characters that are not consistent. Causvid does not have this problem but requires more steps.
|
| 76 |
+
|
| 77 |
Ill try to update this further as things come to mind.
|
| 78 |
|
| 79 |
|