from ctransformers import AutoModelForCausalLM
from fastapi import FastAPI, Form
from pydantic import BaseModel

#Model loading
llm1 = AutoModelForCausalLM.from_pretrained("zephyr-7b-beta.Q4_K_S.gguf",
model_type='mistral',
max_new_tokens = 1096,
threads = 3,
)

llm = AutoModelForCausalLM.from_pretrained("TheBloke/CodeLlama-7B-Instruct-GGUF", 
model_file="codellama-7b-instruct.q4_K_M.gguf",                                            
model_type="codellama", 
gpu_layers=0)



   

#Pydantic object
class validation(BaseModel):
    prompt: str
    
#Fast API
app = FastAPI()

#Zephyr completion
@app.post("/llm_on_cpu")
async def stream(item: validation):
    system_prompt = 'Below is an instruction that describes a task. Write a response that appropriately completes the request.'
    E_INST = "</s>"
    user, assistant = "<|user|>", "<|assistant|>"
    prompt = f"{system_prompt}{E_INST}\n{user}\n{item.prompt.strip()}{E_INST}\n{assistant}\n"
    return llm(prompt)