Update app.py

app.py

@@ -3,12 +3,14 @@ import numpy as np
 import random
 import spaces
 import torch
-from diffusers import DiffusionPipeline
+# Substituindo a importação da DiffusionPipeline
+from custom_pipeline import FLUXPipelineWithIntermediateOutputs 
 
 dtype = torch.bfloat16
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
-pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-schnell", torch_dtype=dtype).to(device)
+# Instanciando a FLUXPipelineWithIntermediateOutputs
+pipe = FLUXPipelineWithIntermediateOutputs.from_pretrained("black-forest-labs/FLUX.1-schnell", torch_dtype=dtype).to(device)
 
 MAX_SEED = np.iinfo(np.int32).max
 MAX_IMAGE_SIZE = 2048
@@ -18,14 +20,19 @@ def infer(prompt, seed=42, randomize_seed=False, width=1024, height=1024, num_in
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
     generator = torch.Generator().manual_seed(seed)
-    image = pipe(
-            prompt = prompt, 
-            width = width,
-            height = height,
-            num_inference_steps = num_inference_steps, 
-            generator = generator,
-            guidance_scale=0.0
-    ).images[0] 
+
+    # Ajustando a chamada da função de inferência
+    for img in pipe.generate_images(
+            prompt=prompt,
+            width=width,
+            height=height,
+            num_inference_steps=num_inference_steps,
+            generator=generator,
+            guidance_scale=0.0,
+            output_type="pil" # Certifique-se que sua pipeline retorna imagens PIL
+    ):
+        image = img # Pegando a primeira imagem do iterador
+
     return image, seed
  
 examples = [
@@ -33,14 +40,12 @@ examples = [
     "a cat holding a sign that says hello world",
     "an anime illustration of a wiener schnitzel",
 ]
-
 css="""
 #col-container {
     margin: 0 auto;
     max-width: 520px;
 }
 """
-
 with gr.Blocks(css=css) as demo:
     
     with gr.Column(elem_id="col-container"):
@@ -111,12 +116,10 @@ with gr.Blocks(css=css) as demo:
             outputs = [result, seed],
             cache_examples="lazy"
         )
-
     gr.on(
         triggers=[run_button.click, prompt.submit],
         fn = infer,
         inputs = [prompt, seed, randomize_seed, width, height, num_inference_steps],
         outputs = [result, seed]
     )
-
 demo.launch()