Update

app.py

@@ -1,7 +1,6 @@
 #!/usr/bin/env python
 
 import os
-import random
 
 import gradio as gr
 import numpy as np
@@ -34,10 +33,22 @@ refiner = DiffusionPipeline.from_pretrained(
 ).to(device)
 
 
-def randomize_seed_fn(seed: int, randomize_seed: bool) -> int:
-    if randomize_seed:
-        seed = random.randint(0, MAX_SEED)  # noqa: S311
-    return seed
+def get_seed(randomize_seed: bool, seed: int) -> int:
+    """Determine and return the random seed to use for model generation or sampling.
+
+    - MAX_SEED is the maximum value for a 32-bit integer (np.iinfo(np.int32).max).
+    - This function is typically used to ensure reproducibility or to introduce randomness in model generation.
+    - The random seed affects the stochastic processes in downstream model inference or sampling.
+
+    Args:
+        randomize_seed (bool): If True, a random seed (an integer in [0, MAX_SEED)) is generated using NumPy's default random number generator. If False, the provided seed argument is returned as-is.
+        seed (int): The seed value to use if randomize_seed is False.
+
+    Returns:
+        int: The selected seed value. If randomize_seed is True, a randomly generated integer; otherwise, the value of the seed argument.
+    """
+    rng = np.random.default_rng()
+    return int(rng.integers(0, MAX_SEED)) if randomize_seed else seed
 
 
 @spaces.GPU
@@ -59,6 +70,36 @@ def generate(
     apply_refiner: bool = False,
     progress: gr.Progress = gr.Progress(track_tqdm=True),  # noqa: ARG001, B008
 ) -> PIL.Image.Image:
+    """Generates an image from a text prompt using the SDXL (Stable Diffusion XL) model.
+
+    This function allows fine-grained control over image generation through prompts,
+    negative prompts, and optional refinement stages.
+
+    Note:
+        All prompt-related inputs (e.g., `prompt`, `negative_prompt`, `prompt_2`, and `negative_prompt_2`)
+        must be written in English for proper model performance.
+
+    Args:
+        prompt (str): Main text prompt used to guide image generation.
+        negative_prompt (str, optional): Text specifying elements to exclude from the image.
+        prompt_2 (str, optional): Secondary prompt for additional guidance. Used only if `use_prompt_2` is True.
+        negative_prompt_2 (str, optional): Secondary negative prompt. Used only if `use_negative_prompt_2` is True.
+        use_negative_prompt (bool, optional): Whether to apply `negative_prompt` during generation.
+        use_prompt_2 (bool, optional): Whether to apply `prompt_2` during generation.
+        use_negative_prompt_2 (bool, optional): Whether to apply `negative_prompt_2` during generation.
+        seed (int, optional): Seed for random number generation. Use 0 to generate a random seed.
+        width (int, optional): Width of the output image in pixels.
+        height (int, optional): Height of the output image in pixels.
+        guidance_scale_base (float, optional): Guidance scale for the base model. Higher values follow the prompt more closely.
+        guidance_scale_refiner (float, optional): Guidance scale for the refiner model.
+        num_inference_steps_base (int, optional): Number of inference steps for the base model.
+        num_inference_steps_refiner (int, optional): Number of inference steps for the refiner model.
+        apply_refiner (bool, optional): Whether to apply the refiner stage after the base image is generated.
+        progress (gr.Progress, optional): Gradio progress object to show progress during generation.
+
+    Returns:
+        PIL.Image.Image: The generated image as a PIL Image object.
+    """
     generator = torch.Generator().manual_seed(seed)
 
     if not use_negative_prompt:
@@ -249,11 +290,10 @@ with gr.Blocks(css_paths="style.css") as demo:
             prompt_2.submit,
             negative_prompt_2.submit,
         ],
-        fn=randomize_seed_fn,
-        inputs=[seed, randomize_seed],
+        fn=get_seed,
+        inputs=[randomize_seed, seed],
         outputs=seed,
         queue=False,
-        api_name=False,
     ).then(
         fn=generate,
         inputs=[
@@ -278,4 +318,4 @@ with gr.Blocks(css_paths="style.css") as demo:
     )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(mcp_server=True)