app.py

from fastapi import FastAPI
from transformers import AutoTokenizer, AutoModelForCausalLM
import uvicorn

tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")
model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-0.5B-Instruct")

app = FastAPI()

@app.get("/")
def root():
    return {"message": "AI API is running"}

@app.post("/chat")
def chat(data: dict):
    msg = data["message"]
    inputs = tokenizer(msg, return_tensors="pt")
    outputs = model.generate(**inputs, max_length=200)
    res = tokenizer.decode(outputs[0], skip_special_tokens=True)
    return {"response": res}

if __name__ == "__main__":
    uvicorn.run(app, host="0.0.0.0", port=7860)