Update app.py

app.py

@@ -1,77 +1,62 @@
-# Thanks: https://huggingface.co/spaces/markmagic/Stable-Diffusion-3/blob/main/app.py
-
-import os
-import random
-import uuid
 
 import gradio as gr
 import numpy as np
-from PIL import Image
-import spaces
+import random
 import torch
-from diffusers import StableDiffusion3Pipeline, DPMSolverMultistepScheduler, AutoencoderKL
+from diffusers import StableDiffusion3Pipeline, SD3Transformer2DModel, FlowMatchEulerDiscreteScheduler
+import spaces
 
-DESCRIPTION = """# 日本語で入力できるStable Diffusion 3"""
+device = "cuda"
+dtype = torch.float16
 
-pipe = StableDiffusion3Pipeline.from_pretrained(
-    "stabilityai/stable-diffusion-3-medium-diffusers",
-    torch_dtype=torch.float16,
-    token=os.getenv("TOKEN")
-)
+repo = "stabilityai/stable-diffusion-3-medium"
+pipe = StableDiffusion3Pipeline.from_pretrained(repo, torch_dtype=torch.float16, revision="refs/pr/26",token=os.environ["TOKEN"]).to(device)
 
-@spaces.GPU()
-def generate(
-    prompt: str,
-    negative_prompt: str = "",
-    seed: int = 0,
-    width: int = 1024,
-    height: int = 1024,
-    guidance_scale: float = 7,
-    num_inference_steps=30,
-    progress=gr.Progress(track_tqdm=True),
-):
-    pipe = pipe.to("cuda")
-    generator = torch.Generator().manual_seed(seed)
-    
-    output = pipe(
-        prompt=prompt,
-        negative_prompt=negative_prompt,
-        width=width,
-        height=height,
-        guidance_scale=guidance_scale,
-        num_inference_steps=num_inference_steps,
-        generator=generator,
-        output_type="pil",
-    ).images
+MAX_SEED = np.iinfo(np.int32).max
+MAX_IMAGE_SIZE = 1344
 
-    return output
+@spaces.GPU
+def infer(prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps, progress=gr.Progress(track_tqdm=True)):
 
+    if randomize_seed:
+        seed = random.randint(0, MAX_SEED)
+        
+    generator = torch.Generator().manual_seed(seed)
+    
+    image = pipe(
+        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
 
 examples = [
-    "A red sofa on top of a white building.",
+    "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k",
+    "An astronaut riding a green horse",
+    "A delicious ceviche cheesecake slice",
 ]
 
-css = '''
-.gradio-container{max-width: 1000px !important}
-h1{text-align:center}
-'''
+css="""
+#col-container {
+    margin: 0 auto;
+    max-width: 580px;
+}
+"""
+
 with gr.Blocks(css=css) as demo:
-    with gr.Row():
-        with gr.Column():
-            gr.HTML(
-            """
-            <h1 style='text-align: center'>
-            日本語で入力できるStable Diffusion 3 Medium
-            </h1>
-            """
-        )
-            gr.HTML(
-                """
-               
-                """
-        )
-    with gr.Group():
+    
+    with gr.Column(elem_id="col-container"):
+        gr.Markdown(f"""
+        # 日本語が入力できる [SD3 Medium](https://huggingface.co/stabilityai/stable-diffusion-3-medium) 
+        """)
+        
         with gr.Row():
+            
             prompt = gr.Text(
                 label="Prompt",
                 show_label=False,
@@ -79,65 +64,74 @@ with gr.Blocks(css=css) as demo:
                 placeholder="Enter your prompt",
                 container=False,
             )
+            
             run_button = gr.Button("Run", scale=0)
-        result = gr.Gallery(label="Result", elem_id="gallery", show_label=False)
-    with gr.Accordion("Advanced options", open=False):
-        with gr.Row():
+        
+        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,
-                value = "deformed, distorted, disfigured, poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, mutated hands and fingers, disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, NSFW",
-                visible=True,
+                placeholder="Enter a negative prompt",
             )
-        seed = gr.Slider(
-            label="Seed",
-            minimum=0,
-            maximum=MAX_SEED,
-            step=1,
-            value=0,
-        )
-
-        steps = gr.Slider(
-            label="Steps",
-            minimum=0,
-            maximum=60,
-            step=1,
-            value=30,
-        )
-        
-        with gr.Row():
-            guidance_scale = gr.Slider(
-                label="Guidance Scale",
-                minimum=0.1,
-                maximum=10,
-                step=0.1,
-                value=7.0,
+            
+            seed = gr.Slider(
+                label="Seed",
+                minimum=0,
+                maximum=MAX_SEED,
+                step=1,
+                value=0,
             )
-
-    gr.Examples(
-        examples=examples,
-        inputs=prompt,
-        outputs=[result],
-        fn=generate,
-        cache_examples=CACHE_EXAMPLES,
-    )
-
+            
+            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+            
+            with gr.Row():
+                
+                width = gr.Slider(
+                    label="Width",
+                    minimum=256,
+                    maximum=MAX_IMAGE_SIZE,
+                    step=64,
+                    value=1024,
+                )
+                
+                height = gr.Slider(
+                    label="Height",
+                    minimum=256,
+                    maximum=MAX_IMAGE_SIZE,
+                    step=64,
+                    value=1024,
+                )
+            
+            with gr.Row():
+                
+                guidance_scale = gr.Slider(
+                    label="Guidance scale",
+                    minimum=0.0,
+                    maximum=10.0,
+                    step=0.1,
+                    value=5.0,
+                )
+                
+                num_inference_steps = gr.Slider(
+                    label="Number of inference steps",
+                    minimum=1,
+                    maximum=50,
+                    step=1,
+                    value=28,
+                )
+        
+        gr.Examples(
+            examples = examples,
+            inputs = [prompt]
+        )
     gr.on(
-        triggers=[
-            prompt.submit,
-            negative_prompt.submit,
-            run_button.click,
-        ],
-        fn=generate,
-        inputs=[
-            prompt,
-            negative_prompt,
-            seed,
-            guidance_scale,
-            steps,
-        ],
-        outputs=[result],
+        triggers=[run_button.click, prompt.submit, negative_prompt.submit],
+        fn = infer,
+        inputs = [prompt, negative_prompt, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
+        outputs = [result, seed]
     )
 
-if __name__ == "__main__":
-    demo.queue().launch()
+demo.launch()