Update README.md

README.md

@@ -57,7 +57,7 @@ You can run the smashed model with these steps:
    - Option 1 - Use command line interface (CLI):
        ```bash
        mkdir SG161222-Realistic_Vision_V1.4-turbo-tiny-green-smashed
-       huggingface-cli download PrunaAI/SG161222-Realistic_Vision_V1.4-turbo-tiny-green-smashed --local-dir SG161222-Realistic_Vision_V1.4-turbo-tiny-green-smashed --local-dir-use-symlinks False
+       huggingface-cli download PrunaAI/FLUX.1-schnell-8bit --local-dir FLUX.1-schnell-8bit --local-dir-use-symlinks False
        ```
    - Option 2 - Use Python:
        ```python
@@ -71,7 +71,7 @@ You can run the smashed model with these steps:
     ```python 
     
     import torch
-    
+
     from optimum.quanto import freeze, qfloat8, quantize
     
     from diffusers import FlowMatchEulerDiscreteScheduler, AutoencoderKL
@@ -104,17 +104,17 @@ You can run the smashed model with these steps:
     )
     pipe.text_encoder_2 = text_encoder_2
     pipe.transformer = transformer
-    pipe.enable_model_cpu_offload()
-    
+    # pipe.enable_model_cpu_offload()
+    pipe.to('cuda')
+    print('done')
     generator = torch.Generator().manual_seed(12345)
-    image = pipe(
+    pipe(
     "a cute apple smiling",
     guidance_scale=0.0,
     num_inference_steps=4,
     max_sequence_length=256,
     generator=torch.Generator("cpu").manual_seed(0)
       ).images[0]
-    image.save("flux-schnell.png")
     ```
 
 ## Configurations