Update app.py

app.py

@@ -36,8 +36,10 @@ processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-larg
 model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")
 processor1 = BlipProcessor.from_pretrained("noamrot/FuseCap")
 model2 = BlipForConditionalGeneration.from_pretrained("noamrot/FuseCap")
-pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-3.5-medium")
+# pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-3.5-medium")
+from diffusers import FluxPipeline
 
+pipe = FluxPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16)
 
 
 
@@ -82,17 +84,26 @@ def generate_caption_and_image(image, f, p, d):
     
         prompt = f"Design a high-quality, stylish clothing item that seamlessly blends the essence of {caption1} and {caption2}. The design should prominently feature {f}{d} and incorporate {p}. The final piece should exude sophistication and creativity, suitable for modern trends while retaining an element of timeless appeal. Ensure the textures and patterns complement each other harmoniously, creating a visually striking yet wearable garment."
 
-
+        image = pipe(
+        prompt,
+        height=500,
+        width=500,
+        guidance_scale=3.5,
+        num_inference_steps=50,
+        max_sequence_length=512,
+        generator=torch.Generator("cpu").manual_seed(0)
+        ).images[0]
+        return image
 
     
     
       
     
-        # Generate image based on the caption
-        generated_image = pipe(prompt).images[0]
-        generated_image1 =pipe(prompt).images[0]
+        # # Generate image based on the caption
+        # generated_image = pipe(prompt).images[0]
+        # generated_image1 =pipe(prompt).images[0]
     
-        return generated_image, generated_image1
+        # return generated_image, generated_image1
     return None, None
 # Gradio UI
 iface = gr.Interface(