Add widget gallery & fix diffusers code
Browse files
README.md
CHANGED
|
@@ -12,10 +12,21 @@ tags:
|
|
| 12 |
base_model: "black-forest-labs/FLUX.1-dev"
|
| 13 |
pipeline_tag: text-to-image
|
| 14 |
instance_prompt: MNALSA
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
---
|
| 16 |
|
| 17 |
# Flux Mona Lisa
|
| 18 |
|
|
|
|
|
|
|
| 19 |
Trained on Replicate using:
|
| 20 |
|
| 21 |
https://replicate.com/ostris/flux-dev-lora-trainer/train
|
|
@@ -31,7 +42,7 @@ You should use `MNALSA` to trigger the image generation.
|
|
| 31 |
from diffusers import AutoPipelineForText2Image
|
| 32 |
import torch
|
| 33 |
|
| 34 |
-
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.
|
| 35 |
pipeline.load_lora_weights('fofr/flux-mona-lisa', weight_name='lora.safetensors')
|
| 36 |
image = pipeline('your prompt').images[0]
|
| 37 |
```
|
|
|
|
| 12 |
base_model: "black-forest-labs/FLUX.1-dev"
|
| 13 |
pipeline_tag: text-to-image
|
| 14 |
instance_prompt: MNALSA
|
| 15 |
+
widget:
|
| 16 |
+
- text: >-
|
| 17 |
+
a photo of MNALSA woman with a coffee in a parisian cafe, 50mm, sharp, award winning portrait photography
|
| 18 |
+
output:
|
| 19 |
+
url: https://replicate.delivery/yhqm/OQPHQ9kRw9ZpIBMub19XY8NxKznMDA50eSEjpx52QP3qoVrJA/out-0.webp
|
| 20 |
+
- text: >-
|
| 21 |
+
a photo of MNALSA woman with pink hair at a rave
|
| 22 |
+
output:
|
| 23 |
+
url: https://replicate.delivery/yhqm/apYK6kZFfZUYRyoJ11NzhHY2YXbrjCHajYIiN9EznGR4qVrJA/out-0.webp
|
| 24 |
---
|
| 25 |
|
| 26 |
# Flux Mona Lisa
|
| 27 |
|
| 28 |
+
<Gallery />
|
| 29 |
+
|
| 30 |
Trained on Replicate using:
|
| 31 |
|
| 32 |
https://replicate.com/ostris/flux-dev-lora-trainer/train
|
|
|
|
| 42 |
from diffusers import AutoPipelineForText2Image
|
| 43 |
import torch
|
| 44 |
|
| 45 |
+
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.bfloat16).to('cuda')
|
| 46 |
pipeline.load_lora_weights('fofr/flux-mona-lisa', weight_name='lora.safetensors')
|
| 47 |
image = pipeline('your prompt').images[0]
|
| 48 |
```
|