Update app.py

app.py

@@ -1,66 +1,77 @@
 import gradio as gr
 import numpy as np
 import random
-import spaces
 import torch
-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
+from diffusers import DiffusionPipeline
+from torch import autocast  # Usando autocast para otimizar operações em float16
 
-dtype = torch.bfloat16
+# Verifica se a GPU está disponível
 device = "cuda" if torch.cuda.is_available() else "cpu"
+model_repo_id = "stabilityai/sdxl-turbo"  # Modelo otimizado para velocidade
 
-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()
+# Usando float16 para otimizar a execução na GPU
+torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32
 
-MAX_SEED = np.iinfo(np.int32).max
-MAX_IMAGE_SIZE = 2048
+# Carregando o modelo com otimizações
+pipe = DiffusionPipeline.from_pretrained(model_repo_id, torch_dtype=torch_dtype).to(device)
 
-pipe.flux_pipe_call_that_returns_an_iterable_of_images = flux_pipe_call_that_returns_an_iterable_of_images.__get__(pipe)
+# Max seed
+MAX_SEED = np.iinfo(np.int32).max
+MAX_IMAGE_SIZE = 512  # Dimensões menores para acelerar
 
-@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)):
+# 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
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
-    generator = torch.Generator().manual_seed(seed)
-    
-    for img in pipe.flux_pipe_call_that_returns_an_iterable_of_images(
+
+    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(
             prompt=prompt,
+            negative_prompt=negative_prompt,
             guidance_scale=guidance_scale,
             num_inference_steps=num_inference_steps,
             width=width,
             height=height,
             generator=generator,
-            output_type="pil",
-            good_vae=good_vae,
-        ):
-            yield img, seed
-    
+        ).images[0]
+
+    return image, seed
+
+
+# Exemplos para o Gradio
 examples = [
-    "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",
+    "Astronaut in a jungle, cold color palette, muted colors, detailed, 2k",
+    "An astronaut riding a green horse",
+    "A delicious ceviche cheesecake slice",
 ]
 
-css="""
+css = """
 #col-container {
     margin: 0 auto;
-    max-width: 520px;
+    max-width: 640px;
 }
 """
 
 with gr.Blocks(css=css) as demo:
-    
     with gr.Column(elem_id="col-container"):
-        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)]
-        """)
-        
+        gr.Markdown(" # Text-to-Image Gradio Template")
+
         with gr.Row():
-            
             prompt = gr.Text(
                 label="Prompt",
                 show_label=False,
@@ -68,13 +79,19 @@ with gr.Blocks(css=css) as demo:
                 placeholder="Enter your prompt",
                 container=False,
             )
-            
-            run_button = gr.Button("Run", scale=0)
-        
+
+            run_button = gr.Button("Run", scale=0, variant="primary")
+
         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,
@@ -82,59 +99,62 @@ 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=1024,
+                    value=512,  # Dimensões reduzidas para otimizar
                 )
-                
+
                 height = gr.Slider(
                     label="Height",
-                    minimum=256,
+                    minimum=576,
                     maximum=MAX_IMAGE_SIZE,
                     step=32,
-                    value=1024,
+                    value=1024,  # Dimensões reduzidas para otimizar
                 )
-            
-            with gr.Row():
 
+            with gr.Row():
                 guidance_scale = gr.Slider(
-                    label="Guidance Scale",
-                    minimum=1,
-                    maximum=15,
+                    label="Guidance scale",
+                    minimum=0.0,
+                    maximum=10.0,
                     step=0.1,
-                    value=3.5,
+                    value=7.5,  # Valor adequado para controle
                 )
-  
+
                 num_inference_steps = gr.Slider(
-                    label="Number of inference steps",
+                    label="Inference steps",
                     minimum=1,
-                    maximum=50,
+                    maximum=30,  # Menos passos para otimizar a velocidade
                     step=1,
-                    value=28,
+                    value=20,  # Um valor equilibrado
                 )
-        
-        gr.Examples(
-            examples = examples,
-            fn = infer,
-            inputs = [prompt],
-            outputs = [result, seed],
-            cache_examples="lazy"
-        )
+
+        gr.Examples(examples=examples, inputs=[prompt])
 
     gr.on(
         triggers=[run_button.click, prompt.submit],
-        fn = infer,
-        inputs = [prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
-        outputs = [result, seed]
+        fn=infer,
+        inputs=[
+            prompt,
+            negative_prompt,
+            seed,
+            randomize_seed,
+            width,
+            height,
+            guidance_scale,
+            num_inference_steps,
+        ],
+        outputs=[result, seed],
     )
 
-demo.launch()
+if __name__ == "__main__":
+    demo.launch()
+