inference api

README.md

@@ -1,12 +1,4 @@
 ---
-title: flux-controlnet-upscaler
-emoji: 🐳
-colorFrom: blue
-colorTo: green
-sdk: gradio
-sdk_version: 5.1.0
-app_file: app.py
-pinned: false
 base_model:
     - black-forest-labs/FLUX.1-dev
 library_name: diffusers
@@ -14,7 +6,7 @@ license: other
 license_name: flux-1-dev-non-commercial-license
 license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
 pipeline_tag: image-to-image
-inference: False
+inference: True
 tags:
     - ControlNet
     - super-resolution

app.py

@@ -1,105 +0,0 @@
-import gradio as gr
-import numpy as np
-import spaces  # Ensure this is correctly imported based on Hugging Face's SDK
-import torch
-from diffusers import FluxControlNetModel, FluxControlNetPipeline
-from diffusers.utils import load_image
-from gradio_imageslider import ImageSlider
-from PIL import Image
-
-# Initialize global variables for the model and pipeline
-controlnet = None
-pipe = None
-
-
-@spaces.GPU  # Decorator to ensure this function runs on GPU
-def initialize_pipeline():
-    global controlnet, pipe
-    if controlnet is None or pipe is None:
-        # Load the ControlNet model with appropriate dtype
-        controlnet = FluxControlNetModel.from_pretrained(
-            "jasperai/Flux.1-dev-Controlnet-Upscaler", torch_dtype=torch.float16
-        )
-
-        # Load the Flux ControlNet Pipeline
-        pipe = FluxControlNetPipeline.from_pretrained(
-            "black-forest-labs/FLUX.1-dev",
-            controlnet=controlnet,
-            torch_dtype=torch.float16,
-        )
-
-        # Move the pipeline to GPU
-        pipe.to("cuda")
-    return pipe
-
-
-def process_image(input_image):
-    try:
-        if input_image is None:
-            raise gr.Error("Please provide an image to upscale.")
-
-        # Convert the input image (numpy array) to PIL Image
-        pil_image = Image.fromarray(input_image)
-
-        # Resize the image by a factor of 4 for upscaling
-        w, h = pil_image.size
-        control_image = pil_image.resize((w * 4, h * 4))
-
-        # Initialize the pipeline (ensures it's loaded on GPU)
-        pipe = initialize_pipeline()
-
-        # Perform upscaling using the Flux pipeline
-        upscaled_image = pipe(
-            prompt="",  # Empty prompt as per your example
-            control_image=control_image,
-            controlnet_conditioning_scale=0.6,
-            num_inference_steps=28,
-            guidance_scale=3.5,
-            height=control_image.size[1],
-            width=control_image.size[0],
-        ).images[0]
-
-        # Convert the upscaled PIL Image to a numpy array
-        result_array = np.array(upscaled_image)
-
-        return [input_image, result_array]
-    except Exception as e:
-        raise gr.Error(f"An error occurred during processing: {e}")
-
-
-# Define the HTML title and description
-title = """<h1 align="center">Flux Upscaler</h1>
-<p><center>Upscales your images by 4x using Flux ControlNet Pipeline</center></p>
-<p><center>
-<a href="https://huggingface.co/jasperai/Flux.1-dev-Controlnet-Upscaler" target="_blank">[ControlNet Model]</a>
-<a href="https://huggingface.co/black-forest-labs/FLUX.1-dev" target="_blank">[Flux Pipeline]</a>
-</center></p>
-<br/>
-<p>This application leverages the Flux ControlNet Pipeline for high-quality image upscaling.</p>
-"""
-
-# Define the Gradio Blocks interface
-with gr.Blocks() as demo:
-
-    gr.HTML(title)
-
-    with gr.Row():
-        with gr.Column(scale=1):
-            input_image = gr.Image(label="Input Image", type="numpy")
-            process_btn = gr.Button(value="Upscale Image", variant="primary")
-        with gr.Column(scale=1):
-            output_slider = ImageSlider(label="Before / After", type="numpy")
-
-    process_btn.click(fn=process_image, inputs=[input_image], outputs=output_slider)
-
-    # Add examples (ensure these images are available in your repository)
-    gr.Examples(
-        examples=["examples/image1.png", "examples/image2.png"],
-        inputs=input_image,
-        outputs=output_slider,
-        fn=process_image,
-        cache_examples=True,
-    )
-
-# Launch the Gradio app
-demo.launch(debug=True)

diffusion_pytorch_model.safetensors

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:2a7ea24d2037ff2aa4d25f8b4ce9fe7e739a2cfe6b9d05106788005d5058c8ca
-size 3583232168

requirements.txt

@@ -1,5 +0,0 @@
-torch>=1.9.0
-diffusers>=0.11.1
-transformers>=4.15.0
-accelerate
-Pillow