add lora weight

app.py

@@ -17,8 +17,8 @@ MAX_IMAGE_SIZE = 2048
 # Initialize the pipeline globally
 pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16).to(device)
 
-@spaces.GPU(duration=190)
-def infer(prompt, lora_model, seed=42, randomize_seed=False, width=1024, height=1024, guidance_scale=5.0, num_inference_steps=28, progress=gr.Progress(track_tqdm=True)):
+@spaces.GPU(duration=300)
+def infer(prompt, lora_model, lora_weight=1.0, seed=42, randomize_seed=False, width=1024, height=1024, guidance_scale=5.0, num_inference_steps=28, progress=gr.Progress(track_tqdm=True)):
     global pipe
     
     # Load LoRA if specified
@@ -39,7 +39,8 @@ def infer(prompt, lora_model, seed=42, randomize_seed=False, width=1024, height=
             height=height,
             num_inference_steps=num_inference_steps, 
             generator=generator,
-            guidance_scale=guidance_scale
+            guidance_scale=guidance_scale,
+            cross_attention_kwargs={"scale": lora_weight}
         ).images[0]
         
         # Unload LoRA weights after generation
@@ -127,6 +128,13 @@ with gr.Blocks(css=css) as demo:
                     step=1,
                     value=28,
                 )
+            lora_weight = gr.Slider(
+                label="LoRA Weight",
+                minimum=0,
+                maximum=2,
+                step=0.01,
+                value=1.0,
+            )
         
         gr.Examples(
             examples=examples,
@@ -139,8 +147,8 @@ with gr.Blocks(css=css) as demo:
     gr.on(
         triggers=[run_button.click, prompt.submit],
         fn=infer,
-        inputs=[prompt, lora_model, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
+        inputs=[prompt, lora_model, lora_weight, seed, randomize_seed, width, height, guidance_scale, num_inference_steps],
         outputs=[result, seed, output_message]
     )
 
-demo.launch()
+demo.launch()