app.py

import torch
from fastapi import FastAPI, Depends
from fastapi.responses import JSONResponse
from authenticator import authenticate_token
from payload_model import PayloadModel
from models import InternVL3
from internvl_utils import internvl_inference


app = FastAPI()

model = InternVL3("OpenGVLab/InternVL3-1B-Instruct")

@app.get("/healthcheck")
def healthcheck():
    return JSONResponse(status_code=200, content={"status": "ok"})

@app.post("/internvl_inference")
async def inference(payload: PayloadModel, token: str = Depends(authenticate_token)):
    try:
        model_response = await internvl_inference(model, payload)
        return JSONResponse(status_code=200, content={"status": "ok", "response": model_response})
    except Exception as e:
        print(f"Error: {e}")
        return JSONResponse(status_code=500, content={"status": "error", "message": str(e)})