How get output from fine tuned llama3 model(trained with alpaca format dataset) in a json format ?

[ApurvPujari]

Is your feature request related to a problem? Please describe.
I have my own gguf file.
I was loading it like this :

from llama_cpp import Llama
llm = Llama(model_path=model_pth,
                    n_ctx=CONTEXT_SIZE,
                    chat_format="alpaca")
model_output = model(user_prompt,max_tokens=max_tokens,temperature=temperature,
      top_p=top_p,echo=False,stop=stop)
print(model_ouput)

but the "model_ouput" is of string type, and its format is not consistant. Thus I cant extract desired output from this string using a common function.

Describe the solution you'd like
I want to know a way, where I can get output in a consistant structured format.

Describe alternatives you've considered
I've tried this, but it doesnt work

llm = Llama(model_path=model_pth,
                    n_ctx=CONTEXT_SIZE,
                    chat_format="alpaca")
llm.create_chat_completion(
    messages=[
        {
            "role": "system",
            "content": "You are a helpful assistant that outputs in JSON.",
        },
        {"role": "user", "content": "Who won the world series in 2020"},
    ],
    response_format={
        "type": "json_object",
    },
    temperature=0.7,
)

Additional context
I am new to this field, any suggestion is appriciated. Thanks !