Upload folder using huggingface_hub
Browse files- .gitattributes +2 -0
- README.md +6 -2
- assets/banner_0_1_0-2.png +3 -0
- assets/banner_0_1_0-4.png +3 -0
.gitattributes
CHANGED
@@ -38,4 +38,6 @@ assets/banner_0_0_8.png filter=lfs diff=lfs merge=lfs -text
|
|
38 |
assets/banner_0_0_9.png filter=lfs diff=lfs merge=lfs -text
|
39 |
assets/banner_0_1_0-0.png filter=lfs diff=lfs merge=lfs -text
|
40 |
assets/banner_0_1_0-1.png filter=lfs diff=lfs merge=lfs -text
|
|
|
41 |
assets/banner_0_1_0-3.png filter=lfs diff=lfs merge=lfs -text
|
|
|
|
38 |
assets/banner_0_0_9.png filter=lfs diff=lfs merge=lfs -text
|
39 |
assets/banner_0_1_0-0.png filter=lfs diff=lfs merge=lfs -text
|
40 |
assets/banner_0_1_0-1.png filter=lfs diff=lfs merge=lfs -text
|
41 |
+
assets/banner_0_1_0-2.png filter=lfs diff=lfs merge=lfs -text
|
42 |
assets/banner_0_1_0-3.png filter=lfs diff=lfs merge=lfs -text
|
43 |
+
assets/banner_0_1_0-4.png filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
@@ -4,7 +4,7 @@ library_name: diffusers
|
|
4 |
pipeline_tag: text-to-image
|
5 |
---
|
6 |
|
7 |
-
<img src="https://huggingface.co/ostris/OpenFLUX.1/resolve/main/assets/banner_0_1_0-
|
8 |
|
9 |
<div style="color: #f0b800;">
|
10 |
|
@@ -20,9 +20,13 @@ After numerous iterations and spending way too much of my own money on compute t
|
|
20 |
|
21 |
This is a fine tune of the [FLUX.1-schnell](https://huggingface.co/black-forest-labs/FLUX.1-schnell) model that has had the distillation trained out of it. Flux Schnell is licensed Apache 2.0, but it is a distilled model, meaning you cannot fine-tune it. However, it is an amazing model that can generate amazing images in 1-4 steps. This is an attempt to remove the distillation to create an open source, permissivle licensed model that can be fine tuned.
|
22 |
|
|
|
|
|
|
|
23 |
## How to Use
|
24 |
|
25 |
Since the distillation has been fine tuned out of the model, it uses classic CFG. Since it requires CFG, it will require a different pipeline than the original FLUX.1 schnell and dev models. This pipeline can be found in open_flux_pipeline.py in this repo. I will be adding example code in the next few days, but for now, a cfg of 3.5 seems to work well.
|
26 |
|
|
|
27 |
|
28 |
-
<img src="https://huggingface.co/ostris/OpenFLUX.1/resolve/main/assets/banner_0_1_0-
|
|
|
4 |
pipeline_tag: text-to-image
|
5 |
---
|
6 |
|
7 |
+
<img src="https://huggingface.co/ostris/OpenFLUX.1/resolve/main/assets/banner_0_1_0-2.png" style="max-width: 100%;">
|
8 |
|
9 |
<div style="color: #f0b800;">
|
10 |
|
|
|
20 |
|
21 |
This is a fine tune of the [FLUX.1-schnell](https://huggingface.co/black-forest-labs/FLUX.1-schnell) model that has had the distillation trained out of it. Flux Schnell is licensed Apache 2.0, but it is a distilled model, meaning you cannot fine-tune it. However, it is an amazing model that can generate amazing images in 1-4 steps. This is an attempt to remove the distillation to create an open source, permissivle licensed model that can be fine tuned.
|
22 |
|
23 |
+
<img src="https://huggingface.co/ostris/OpenFLUX.1/resolve/main/assets/banner_0_1_0-1.png" style="max-width: 100%;">
|
24 |
+
|
25 |
+
|
26 |
## How to Use
|
27 |
|
28 |
Since the distillation has been fine tuned out of the model, it uses classic CFG. Since it requires CFG, it will require a different pipeline than the original FLUX.1 schnell and dev models. This pipeline can be found in open_flux_pipeline.py in this repo. I will be adding example code in the next few days, but for now, a cfg of 3.5 seems to work well.
|
29 |
|
30 |
+
<img src="https://huggingface.co/ostris/OpenFLUX.1/resolve/main/assets/banner_0_1_0-0.png" style="max-width: 100%;">
|
31 |
|
32 |
+
<img src="https://huggingface.co/ostris/OpenFLUX.1/resolve/main/assets/banner_0_1_0-4.png" style="max-width: 100%;">
|
assets/banner_0_1_0-2.png
ADDED
Git LFS Details
|
assets/banner_0_1_0-4.png
ADDED
Git LFS Details
|