Instructions to use SeanZheng/Distill-WAM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use SeanZheng/Distill-WAM with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("SeanZheng/Distill-WAM", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
galaxea-a1: README points to the teacher folder
Browse files- galaxea-a1/README.md +3 -0
galaxea-a1/README.md
CHANGED
|
@@ -18,3 +18,6 @@ Deploy env: `VA_CKPT_DIR=<dir> VIDEO_NFE=1 ACTION_NFE=1 VIDEO_GS=1`, config `gal
|
|
| 18 |
Cameras: `observation.images.front` (640×480 cropped to 480×480 with crop_xywh=[103,0,480,480]) and
|
| 19 |
`observation.images.wrist` (480×640); 30 Hz; prompts `put the red block on the red plate` /
|
| 20 |
`put the banana on the red plate`.
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
Cameras: `observation.images.front` (640×480 cropped to 480×480 with crop_xywh=[103,0,480,480]) and
|
| 19 |
`observation.images.wrist` (480×640); 30 Hz; prompts `put the red block on the red plate` /
|
| 20 |
`put the banana on the red plate`.
|
| 21 |
+
|
| 22 |
+
Teacher (20-step video / 50-step action diffusion, reference only): `galaxea-a1-teacher/` in this repo,
|
| 23 |
+
deploy with `VIDEO_NFE=20 ACTION_NFE=50 VIDEO_GS=5`.
|