Update app.py

app.py

@@ -1,77 +1,66 @@
 import gradio as gr
 import numpy as np
 import random
+import spaces
 import torch
-from diffusers import DiffusionPipeline
-from torch import autocast  # Usando autocast para otimizar operações em float16
+from diffusers import  DiffusionPipeline, FlowMatchEulerDiscreteScheduler, AutoencoderTiny, AutoencoderKL
+from transformers import CLIPTextModel, CLIPTokenizer,T5EncoderModel, T5TokenizerFast
+from live_preview_helpers import calculate_shift, retrieve_timesteps, flux_pipe_call_that_returns_an_iterable_of_images
 
-# Verifica se a GPU está disponível
+dtype = torch.bfloat16
 device = "cuda" if torch.cuda.is_available() else "cpu"
-model_repo_id = "stabilityai/sdxl-turbo"  # Modelo otimizado para velocidade
 
-# Usando float16 para otimizar a execução na GPU
-torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
+taef1 = AutoencoderTiny.from_pretrained("madebyollin/taef1", torch_dtype=dtype).to(device)
+good_vae = AutoencoderKL.from_pretrained("black-forest-labs/FLUX.1-dev", subfolder="vae", torch_dtype=dtype).to(device)
+pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=dtype, vae=taef1).to(device)
+torch.cuda.empty_cache()
 
-# Carregando o modelo com otimizações
-pipe = DiffusionPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype).to(device)
-
-# Max seed
 MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 512  # Dimensões menores para acelerar
+MAX_IMAGE_SIZE = 2048
+
+pipe.flux_pipe_call_that_returns_an_iterable_of_images = flux_pipe_call_that_returns_an_iterable_of_images.__get__(pipe)
 
-# Função de inferência otimizada
-def infer(
-    prompt,
-    negative_prompt,
-    seed,
-    randomize_seed,
-    width,
-    height,
-    guidance_scale,
-    num_inference_steps,
-    progress=gr.Progress(track_tqdm=True),
-):
-    # Randomiza a semente, se necessário
+@spaces.GPU(duration=75)
+def infer(prompt, seed=42, randomize_seed=False, width=1024, height=1024, guidance_scale=3.5, num_inference_steps=28, progress=gr.Progress(track_tqdm=True)):
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
-
-    generator = torch.Generator(device).manual_seed(seed)
-
-    # Usando autocast para acelerar o cálculo com float16 em GPUs
-    with autocast("cuda"):
-        # Geração da imagem com um número reduzido de passos (para acelerar)
-        image = pipe(
+    generator = torch.Generator().manual_seed(seed)
+    
+    for img in pipe.flux_pipe_call_that_returns_an_iterable_of_images(
             prompt=prompt,
-            negative_prompt=negative_prompt,
             guidance_scale=guidance_scale,
             num_inference_steps=num_inference_steps,
             width=width,
             height=height,
             generator=generator,
-        ).images[0]
-
-    return image, seed
-
-
-# Exemplos para o Gradio
+            output_type="pil",
+            good_vae=good_vae,
+        ):
+            yield img, seed
+    
 examples = [
-    "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
-    "An astronaut riding a green horse",
-    "A delicious ceviche cheesecake slice",
+    "a tiny astronaut hatching from an egg on the moon",
+    "a cat holding a sign that says hello world",
+    "an anime illustration of a wiener schnitzel",
 ]
 
-css = """
+css="""
 #col-container {
     margin: 0 auto;
-    max-width: 640px;
+    max-width: 520px;
 }
 """
 
 with gr.Blocks(css=css) as demo:
+    
     with gr.Column(elem_id="col-container"):
-        gr.Markdown(" # Text-to-Image Gradio Template")
-
+        gr.Markdown(f"""# FLUX.1 [dev]
+12B param rectified flow transformer guidance-distilled from [FLUX.1 [pro]](https://blackforestlabs.ai/)  
+[[non-commercial license](https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md)] [[blog](https://blackforestlabs.ai/announcing-black-forest-labs/)] [[model](https://huggingface.co/black-forest-labs/FLUX.1-dev)]
+        """)
+        
         with gr.Row():
+            
             prompt = gr.Text(
                 label="Prompt",
                 show_label=False,
@@ -79,19 +68,13 @@ with gr.Blocks(css=css) as demo:
                 placeholder="Enter your prompt",
                 container=False,
             )
-
-            run_button = gr.Button("Run", scale=0, variant="primary")
-
+            
+            run_button = gr.Button("Run", scale=0)
+        
         result = gr.Image(label="Result", show_label=False)
-
+        
         with gr.Accordion("Advanced Settings", open=False):
-            negative_prompt = gr.Text(
-                label="Negative prompt",
-                max_lines=1,
-                placeholder="Enter a negative prompt",
-                visible=False,
-            )
-
+            
             seed = gr.Slider(
                 label="Seed",
                 minimum=0,
@@ -99,61 +82,59 @@ with gr.Blocks(css=css) as demo:
                 step=1,
                 value=0,
             )
-
+            
             randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
-
+            
             with gr.Row():
+                
                 width = gr.Slider(
                     label="Width",
                     minimum=256,
                     maximum=MAX_IMAGE_SIZE,
                     step=32,
-                    value=512,  # Dimensões reduzidas para otimizar
+                    value=1024,
                 )
-
+                
                 height = gr.Slider(
                     label="Height",
                     minimum=256,
                     maximum=MAX_IMAGE_SIZE,
                     step=32,
-                    value=512,  # Dimensões reduzidas para otimizar
+                    value=1024,
                 )
-
+            
             with gr.Row():
+
                 guidance_scale = gr.Slider(
-                    label="Guidance scale",
-                    minimum=0.0,
-                    maximum=10.0,
+                    label="Guidance Scale",
+                    minimum=1,
+                    maximum=15,
                     step=0.1,
-                    value=7.5,  # Valor adequado para controle
+                    value=3.5,
                 )
-
+  
                 num_inference_steps = gr.Slider(
-                    label="Inference steps",
+                    label="Number of inference steps",
                     minimum=1,
-                    maximum=30,  # Menos passos para otimizar a velocidade
+                    maximum=50,
                     step=1,
-                    value=20,  # Um valor equilibrado
+                    value=28,
                 )
-
-        gr.Examples(examples=examples, inputs=[prompt])
+        
+        gr.Examples(
+            examples = examples,
+            fn = infer,
+            inputs = [prompt],
+            outputs = [result, seed],
+            cache_examples="lazy"
+        )
 
     gr.on(
         triggers=[run_button.click, prompt.submit],
-        fn=infer,
-        inputs=[
-            prompt,
-            negative_prompt,
-            seed,
-            randomize_seed,
-            width,
-            height,
-            guidance_scale,
-            num_inference_steps,
-        ],
-        outputs=[result, seed],
+        fn = infer,
+        inputs = [prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
+        outputs = [result, seed]
     )
 
-if __name__ == "__main__":
-    demo.launch()
+demo.launch()