app.py

from fastapi import FastAPI
from fastapi.responses import JSONResponse
from diffusers import DiffusionPipeline
import torch
import base64
from io import BytesIO

# create fastapi instance
app = FastAPI()

# Detect device
device = "cuda" if torch.cuda.is_available() else "cpu"

# Load the correct pipeline
pipe = DiffusionPipeline.from_pretrained(
    "segmind/SSD-1B",
    torch_dtype=torch.float16 if device == "cuda" else torch.float32,
    use_safetensors=True,
)
pipe = pipe.to(device)

# Optional: reduce VRAM usage
pipe.enable_attention_slicing()

@app.get("/")
def home():
    return {"message": "Segmind SSD-1B API running"}

# Define function to generate image from text prompt
@app.post("/generate-image/")
def generate_image(prompt: str, negative_prompt: str = None):
    # Run inference
    image = pipe(prompt=prompt, negative_prompt=negative_prompt).images[0]
    
    # Convert to base64 string
    buffered = BytesIO()
    image.save(buffered, format="PNG")
    img_str = base64.b64encode(buffered.getvalue()).decode("utf-8")

    return JSONResponse(content={"image_base64": img_str})