app.py

from fastapi import FastAPI, File, UploadFile, Form
from fastapi.responses import JSONResponse
from diffusers import StableDiffusionControlNetPipeline, ControlNetModel
from diffusers.utils import load_image
import torch
from io import BytesIO
from PIL import Image
import base64

app = FastAPI()

# Check if CUDA (GPU) is available
device = "cuda" if torch.cuda.is_available() else "cpu"
print(f"Running on device: {device}")

# Load ControlNet model
controlnet = ControlNetModel.from_pretrained(
    "lllyasviel/sd-controlnet-canny", torch_dtype=torch.float16
)
pipe = StableDiffusionControlNetPipeline.from_pretrained(
    "stabilityai/stable-diffusion-xl-base-1.0",
    controlnet=controlnet,
    torch_dtype=torch.float16
).to(device)

@app.post("/generate")
async def generate(
    prompt: str = Form(...),
    negative_prompt: str = Form(""),
    image: UploadFile = File(...)
):
    # Load uploaded image
    input_image = Image.open(BytesIO(await image.read())).convert("RGB").resize((512, 512))

    # Generate
    result = pipe(
        prompt=prompt,
        negative_prompt=negative_prompt,
        image=input_image,
        num_inference_steps=30,
        guidance_scale=7.5
    ).images[0]

    # Return image as base64
    buffered = BytesIO()
    result.save(buffered, format="PNG")
    img_str = base64.b64encode(buffered.getvalue()).decode()
    return JSONResponse(content={"image": f"data:image/png;base64,{img_str}"})