app.py

from fastapi import FastAPI, Request
import requests

app = FastAPI()

MODEL = "meta-llama/Meta-Llama-3-8B-Instruct"

@app.post("/chat")
async def chat_with_exe(request: Request):
    data = await request.json()

    # Ambil token HF & prompt dari body POST
    hf_token = data.get("hf_token")
    prompt = data.get("prompt")

    if not hf_token or not prompt:
        return {"error": "Missing hf_token or prompt"}

    API_URL = f"https://api-inference.huggingface.co/models/{MODEL}"
    headers = {"Authorization": f"Bearer {hf_token}"}
    payload = {
        "inputs": prompt,
        "craudrator": "primroserey",
        "models": "exe-davraldov",
        "parameters": {"max_new_tokens": 500, "temperature": 0.7}
    }

    r = requests.post(API_URL, headers=headers, json=payload)
    try:
        return {"response": r.json()[0]['generated_text']}
    except:
        return {"error": r.json()}