Update README.md
Browse files
README.md
CHANGED
|
@@ -1,10 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
| 1 |
diffusers formation for mochi-1-preview model.
|
| 2 |
|
| 3 |
It was create by scripts: https://github.com/huggingface/diffusers/blob/mochi/scripts/convert_mochi_to_diffusers.py
|
| 4 |
|
| 5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
|
| 7 |
|
| 8 |
---
|
| 9 |
license: apache-2.0
|
| 10 |
-
---
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
---
|
| 4 |
diffusers formation for mochi-1-preview model.
|
| 5 |
|
| 6 |
It was create by scripts: https://github.com/huggingface/diffusers/blob/mochi/scripts/convert_mochi_to_diffusers.py
|
| 7 |
|
| 8 |
+
The model can be directly load with from pretrained with mochi branch: https://github.com/huggingface/diffusers/tree/mochi
|
| 9 |
+
|
| 10 |
+
```bash
|
| 11 |
+
from diffusers import MochiPipeline
|
| 12 |
+
from diffusers.utils import export_to_video
|
| 13 |
+
|
| 14 |
+
pipe = MochiPipeline.from_pretrained(path)
|
| 15 |
+
pipe.to("cuda")
|
| 16 |
+
prompt = "Close-up of a chameleon's eye, with its scaly skin changing color. Ultra high resolution 4k."
|
| 17 |
+
frames = pipe(prompt, num_inference_steps=28, guidance_scale=3.5).frames[0]
|
| 18 |
+
|
| 19 |
+
export_to_video(frames, "mochi.mp4")
|
| 20 |
+
```
|
| 21 |
+
|
| 22 |
|
| 23 |
|
| 24 |
---
|
| 25 |
license: apache-2.0
|
| 26 |
+
---
|