main.py

from fastapi import FastAPI
from fastapi.staticfiles import StaticFiles
from fastapi.responses import FileResponse

import subprocess
import os
import json
import uuid

import torch
from diffusers import (
    StableDiffusionPipeline,
    DPMSolverMultistepScheduler,
    EulerDiscreteScheduler,
)

app = FastAPI()


@app.get("/generate")
def generate_image(prompt, model):
    torch.cuda.empty_cache()

    modelArray = model.split(",")
    modelName = modelArray[0]
    modelVersion = modelArray[1]

    pipeline = StableDiffusionPipeline.from_pretrained(
        str(modelName), torch_dtype=torch.float16
    )
    pipeline.scheduler = EulerDiscreteScheduler.from_config(pipeline.scheduler.config)
    pipeline = pipeline.to("cuda")

    image = pipeline(prompt, num_inference_steps=50, height=512, width=512).images[0]

    filename = str(uuid.uuid4()) + ".jpg"
    image.save(filename)

    assertion = {
        "assertions": [
            {
                "label": "com.truepic.custom.ai",
                "data": {
                    "model_name": modelName,
                    "model_version": modelVersion,
                    "prompt": prompt,
                },
            }
        ]
    }

    json_object = json.dumps(assertion)

    subprocess.check_output(
        [
            "./truepic",
            "sign",
            filename,
            "--assertions-inline",
            json_object,
            "--output",
            (os.getcwd() + "/static/" + filename),
        ]
    )

    return {"response": filename}


app.mount("/", StaticFiles(directory="static", html=True), name="static")


@app.get("/")
def index() -> FileResponse:
    return FileResponse(path="/app/static/index.html", media_type="text/html")