Update app.py

app.py

@@ -13,8 +13,11 @@ else:
 
 # Initialize the base model and specific LoRA
 base_model = "black-forest-labs/FLUX.1-dev"
-pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.float32)
-pipe.to("gpu")
+pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.float16)
+
+# Check if CUDA is available and move the model to GPU if possible
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+pipe = pipe.to(device)
 
 lora_repo = "sagar007/sagar_flux"
 trigger_word = "sagar"
@@ -25,10 +28,8 @@ MAX_SEED = 2**32-1
 def run_lora(prompt, cfg_scale, steps, randomize_seed, seed, width, height, lora_scale, progress=gr.Progress(track_tqdm=True)):
     if randomize_seed:
         seed = random.randint(0, MAX_SEED)
-    generator = torch.Generator(device="gpu").manual_seed(seed)
-
-    progress(0, "Starting image generation (this may take a while on CPU)...")
-
+    generator = torch.Generator(device=device).manual_seed(seed)
+    progress(0, f"Starting image generation (using {device})...")
     image = pipe(
         prompt=f"{prompt} {trigger_word}",
         num_inference_steps=steps,
@@ -38,12 +39,34 @@ def run_lora(prompt, cfg_scale, steps, randomize_seed, seed, width, height, lora
         generator=generator,
         cross_attention_kwargs={"scale": lora_scale},
     ).images[0]
-
     progress(100, "Completed!")
-
     return image, seed
 
-# (Rest of the Gradio interface code remains the same)
+# Gradio interface setup
+with gr.Blocks() as app:
+    gr.Markdown("# Text-to-Image Generation with LoRA")
+    with gr.Row():
+        with gr.Column():
+            prompt = gr.Textbox(label="Prompt")
+            run_button = gr.Button("Generate")
+        with gr.Column():
+            result = gr.Image(label="Result")
+    with gr.Row():
+        cfg_scale = gr.Slider(minimum=1, maximum=20, value=7, step=0.1, label="CFG Scale")
+        steps = gr.Slider(minimum=1, maximum=100, value=50, step=1, label="Steps")
+    with gr.Row():
+        width = gr.Slider(minimum=128, maximum=1024, value=512, step=64, label="Width")
+        height = gr.Slider(minimum=128, maximum=1024, value=512, step=64, label="Height")
+    with gr.Row():
+        seed = gr.Number(label="Seed", precision=0)
+        randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
+    lora_scale = gr.Slider(minimum=0, maximum=1, value=0.75, step=0.01, label="LoRA Scale")
+    
+    run_button.click(
+        run_lora,
+        inputs=[prompt, cfg_scale, steps, randomize_seed, seed, width, height, lora_scale],
+        outputs=[result, seed]
+    )
 
 # Launch the app
 app.launch()