from langchain.llms import LlamaCpp
from langchain.callbacks.manager import CallbackManager
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
import gradio as gr
import re
import os



n_gpu_layers = 40 # Change this value based on your model and your GPU VRAM pool.
n_batch = 512 # Should be between 1 and n_ctx, consider the amount of VRAM in your GPU.
n_ctx=2048

callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])
path = "persian_llama_7b.Q8_K_M.gguf"

llm = LlamaCpp(
    model_path= path,
    n_gpu_layers=n_gpu_layers, n_batch=n_batch,
    callback_manager=callback_manager,
    verbose=True,
    n_ctx=n_ctx,
    temperature=0.2,
    max_tokens=200,
    top_p=1,
)

prompt = """Below is an instruction that describes a task.
Write a response that appropriately completes the request.\n\n
### Instruction:\n\n{}\n\n\n### Response:\n\n\n"""
def generate_output(text):
    result = ""
    for s in llm.stream(prompt.format(text)):
        result += s
        yield result


def clear():
    return "", ""

with gr.Blocks(theme=gr.themes.Soft()) as demo:
    with gr.Row():
        inputs=gr.Textbox(label="ورودی",placeholder="سوال خود را وارد کنید",rtl=True)
        
    with gr.Row():
        submit_btn= gr.Button("ارسال", variant="primary")
        clear_btn = gr.ClearButton(value="پاک کردن", variant="secondary")
    with gr.Row():
        outputs=gr.Textbox(label="خروجی",rtl=True)
    submit_btn.click(fn=generate_output,
                    inputs= [inputs],
                    outputs= [outputs])
    clear_btn.click(fn=clear, inputs=[], outputs=[inputs, outputs])

    
demo.launch(server_name='0.0.0.0',share=True)