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

device = "cuda" # the device to load the model onto

model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen2-72B-Instruct",
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2-72B-Instruct")

class ChatRequest(BaseModel):
    prompt: str

app = FastAPI()

@app.get("/")
def greet_json():
    return {"Hello": "World!"}

@app.post("/generate_chat")
def generateAi(request: ChatRequest):
    messages = [
        {"role": "system", "content": "You are a Mandarin language learning assistant that only answers in Mandarin."},
        {"role": "user", "content": request.prompt}
    ]
    text = tokenizer.apply_chat_template(
        messages,
        tokenize=False,
        add_generation_prompt=True
    )
    model_inputs = tokenizer([text], return_tensors="pt").to(device)

    generated_ids = model.generate(
        model_inputs.input_ids,
        max_new_tokens=512
    )
    generated_ids = [
        output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
    ]

    response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
    
    return {"answer": "Hello"}