Update README.md
Browse files
README.md
CHANGED
|
@@ -8,9 +8,10 @@ Similar to our sd-flow-alpha model, but hopefully better trained,
|
|
| 8 |
AND this time it has SDXL vae bolted on.
|
| 9 |
|
| 10 |
Started with our recent [xlsd32-beta1 model](https://huggingface.co/opendiffusionai/xlsd32-beta1)
|
|
|
|
| 11 |
|
| 12 |
Replaced noise scheduler with FlowMatchEulerDiscrete,
|
| 13 |
-
then fine tuned.
|
| 14 |
|
| 15 |
Around 160k images, "high res" (1536px+), square
|
| 16 |
|
|
|
|
| 8 |
AND this time it has SDXL vae bolted on.
|
| 9 |
|
| 10 |
Started with our recent [xlsd32-beta1 model](https://huggingface.co/opendiffusionai/xlsd32-beta1)
|
| 11 |
+
which is already retrained for XL vae, so that part was out of the way.
|
| 12 |
|
| 13 |
Replaced noise scheduler with FlowMatchEulerDiscrete,
|
| 14 |
+
then fine tuned, single run.
|
| 15 |
|
| 16 |
Around 160k images, "high res" (1536px+), square
|
| 17 |
|