multimodalart HF staff commited on
Commit
d2d1035
1 Parent(s): 399ea0a

Add diffusers to docs

Browse files
Files changed (1) hide show
  1. README.md +30 -0
README.md CHANGED
@@ -31,6 +31,36 @@ The FLUX.1 models are also available via API from the following sources
31
  ## ComfyUI
32
  `FLUX.1 [schnell]` is also available in [Comfy UI](https://github.com/comfyanonymous/ComfyUI) for local inference with a node-based workflow.
33
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
34
  ---
35
  # Limitations
36
  - This model is not intended or able to provide factual information.
 
31
  ## ComfyUI
32
  `FLUX.1 [schnell]` is also available in [Comfy UI](https://github.com/comfyanonymous/ComfyUI) for local inference with a node-based workflow.
33
 
34
+ ## Diffusers
35
+ To use `FLUX.1 [schnell]` with the 🧨 diffusers python library, first install or upgrade diffusers
36
+
37
+ ```shell
38
+ pip install git+https://github.com/huggingface/diffusers.git
39
+ ```
40
+
41
+ Then you can use `FluxPipeline` to run the model
42
+
43
+ ```python
44
+ import torch
45
+ from diffusers import FluxPipeline
46
+
47
+ pipe = FluxPipeline.from_pretrained("black-forest-labs/FLUX.1-schnell", torch_dtype=torch.bfloat16)
48
+ pipe.enable_model_cpu_offload() #save some VRAM by offloading the model to CPU. Remove this if you have enough GPU power
49
+
50
+ prompt = "A cat holding a sign that says hello world"
51
+ image = pipe(
52
+ prompt,
53
+ guidance_scale=0.0,
54
+ output_type="pil",
55
+ num_inference_steps=4,
56
+ max_sequence_length=256,
57
+ generator=torch.Generator("cpu").manual_seed(0)
58
+ ).images[0]
59
+ image.save("flux-schnell.png")
60
+ ```
61
+
62
+ To learn mode check out the [diffusers](https://huggingface.co/docs/diffusers/main/en/api/pipelines/flux) documentation
63
+
64
  ---
65
  # Limitations
66
  - This model is not intended or able to provide factual information.