app.py

from fastapi import FastAPI
from pydantic import BaseModel
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
import uvicorn

MODEL_NAME = "TinyLlama/TinyLlama-1.1B-Chat-v1.0"

tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
model = AutoModelForCausalLM.from_pretrained(MODEL_NAME)
device = "cuda" if torch.cuda.is_available() else "cpu"
model.to(device)

app = FastAPI()

class ChatReq(BaseModel):
    message: str

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

@app.post("/chat")
async def chat(data: ChatReq):
    inputs = tokenizer(data.message, return_tensors="pt").to(device)
    with torch.no_grad():
        outputs = model.generate(
            **inputs,
            max_new_tokens=128,
            do_sample=True,
            temperature=0.7,
            pad_token_id=tokenizer.eos_token_id,
        )
    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)