Create app.py

app.py

@@ -0,0 +1,46 @@
+import gradio as gr
+import torch
+import spaces
+from diffusers import FluxPipeline
+from safetensors.torch import load_file
+
+# Load the Flux Dev model
+model_id = "black-forest-labs/FLUX.1-dev"
+pipe = FluxPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
+pipe.to("cuda")
+
+# Load the LoRA weights
+lora_path = "MegaTronX/SuicideGirl-FLUX"
+lora_weights = load_file(lora_path)
+
+# Apply LoRA weights to the model
+pipe.unet.load_attn_procs(lora_weights)
+
+@spaces.GPU
+def generate_image(prompt, negative_prompt, guidance_scale, num_inference_steps, lora_scale):
+    with torch.inference_mode():
+        image = pipe(
+            prompt=prompt,
+            negative_prompt=negative_prompt,
+            num_inference_steps=num_inference_steps,
+            guidance_scale=guidance_scale,
+            cross_attention_kwargs={"scale": lora_scale},
+        ).images[0]
+    return image
+
+# Create the Gradio interface
+iface = gr.Interface(
+    fn=generate_image,
+    inputs=[
+        gr.Textbox(label="Prompt"),
+        gr.Textbox(label="Negative Prompt"),
+        gr.Slider(1, 20, value=7.5, label="Guidance Scale"),
+        gr.Slider(1, 100, value=50, step=1, label="Number of Inference Steps"),
+        gr.Slider(0, 1, value=0.75, label="LoRA Scale"),
+    ],
+    outputs=gr.Image(type="pil"),
+    title="Flux Dev with Custom LoRA Image Generator",
+    description="Generate images using Flux Dev model with a custom LoRA trained on Civitai",
+)
+
+iface.launch()