Add diffusers code example
Browse files
README.md
CHANGED
@@ -24,6 +24,30 @@ The model has been mostly developed by Founder and CEO of Mann-E, [Muhammadreza
|
|
24 |
- Scale: 768x768 and 832x832 are just fine. Higher isn't tested. For 16:9 just try 1080x608
|
25 |
- Sampler : DPM++ SDE Karras
|
26 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
27 |
## Additional Notes
|
28 |
|
29 |
- SDXL 1.0 LoRas are working just fine with the model.
|
|
|
24 |
- Scale: 768x768 and 832x832 are just fine. Higher isn't tested. For 16:9 just try 1080x608
|
25 |
- Sampler : DPM++ SDE Karras
|
26 |
|
27 |
+
## Use it with diffusers
|
28 |
+
|
29 |
+
```py
|
30 |
+
from diffusers import DiffusionPipeline, DPMSolverSinglestepScheduler
|
31 |
+
import torch
|
32 |
+
|
33 |
+
pipe = DiffusionPipeline.from_pretrained(
|
34 |
+
"mann-e/Mann-E_Dreams", torch_dtype=torch.float16
|
35 |
+
).to("cuda")
|
36 |
+
|
37 |
+
#This is equivalent to DPM++ SDE Karras, as noted in https://huggingface.co/docs/diffusers/main/en/api/schedulers/overview
|
38 |
+
pipe.scheduler = DPMSolverSinglestepScheduler.from_config(pipe.scheduler.config, use_karras_sigmas=True)
|
39 |
+
|
40 |
+
image = pipe(
|
41 |
+
prompt="a cat in a bustling middle eastern city",
|
42 |
+
num_inference_steps=8,
|
43 |
+
guidance_scale=3,
|
44 |
+
width=768,
|
45 |
+
height=768,
|
46 |
+
clip_skip=1
|
47 |
+
).images[0]
|
48 |
+
image
|
49 |
+
```
|
50 |
+
|
51 |
## Additional Notes
|
52 |
|
53 |
- SDXL 1.0 LoRas are working just fine with the model.
|