chain_app.py

import chainlit as cl
from gradio_client import Client
from openai import OpenAI
from groq import Groq
import requests
from chainlit.input_widget import Select, Slider
import os
import cohere 
from huggingface_hub import InferenceClient


hf_token = os.environ.get("HF_TOKEN")
hf_token_llama_3_1 = os.environ.get('HF_TOKEN_FOR_31')
openai_api_key = os.environ.get('OPENAI_API_KEY')
groq_api_key = os.environ.get('GROQ_API_KEY')
cohere_api_key = os.environ.get('COHERE_API_KEY')


hf_text_client = Client("Artin2009/text-generation", hf_token=hf_token)
# hf_image_client = Client('Artin2009/image-generation')
openai_client = OpenAI(api_key=openai_api_key)
groq_client = Groq(api_key=groq_api_key)

co = cohere.Client(
  api_key=cohere_api_key,
) 

# API_URL = "https://api-inference.huggingface.co/models/PartAI/TookaBERT-Large"
# headers = {"Authorization": f"Bearer {hf_token}"}


# def query(payload):
# 	response = requests.post(API_URL, headers=headers, json=payload)
# 	return response.json()

@cl.set_chat_profiles
async def chat_profile():
    return [
        cl.ChatProfile(
            name="None",
            markdown_description="None",
        ),
        cl.ChatProfile(
            name="neural-brain-AI",
            markdown_description="The main model of neural brain",
        ),
        cl.ChatProfile(
            name="Dorna-AI",
            markdown_description="One of the open-sourced models that neural brain team fine-tuned",
        ),
        # cl.ChatProfile(
        #     name='Image-Generation',
        #     markdown_description='Our image generation model, has a performance like midjourney',
        # ),
        cl.ChatProfile(
            name="gpt4-o-mini",
            markdown_description="The best state of the art openai model",
        ),
        cl.ChatProfile(
            name="GPT-4",
            markdown_description="OpenAI's GPT-4 model",
        ),
        cl.ChatProfile(
            name="gpt-3.5-turbo",
            markdown_description="OpenAI's GPT-3.5 Turbo model",
        ),
        # cl.ChatProfile(
        #     name="GPT-3.5-turbo-0125",
        #     markdown_description="OpenAI's GPT-3.5 Turbo 0125 model",
        # ),
        cl.ChatProfile(
            name="gpt-3.5-turbo-1106",
            markdown_description="OpenAI's GPT-3.5 Turbo 1106 model",
        ),
        # cl.ChatProfile(
        #     name="davinci-002",
        #     markdown_description="OpenAI's Davinci-002 model",
        # ),
        cl.ChatProfile(
            name="TTS",
            markdown_description="OpenAI's Text-to-Speech model",
        ),
        cl.ChatProfile(
            name="Qwen2-57B",
            markdown_description="Qwen second generation model with 57B parameters",
        ),
        cl.ChatProfile(
            name="Qwen2-7B",
            markdown_description="Qwen second generation model with 7B parameters",
        ),
        cl.ChatProfile(
            name="Qwen2-1.5B",
            markdown_description="Qwen second generation model with 1.5B parameters",
        ),
        cl.ChatProfile(
            name="Qwen2-0.5B",
            markdown_description="Qwen second generation model with 1.5B parameters",
        ),
        cl.ChatProfile(
            name="Qwen1.5-110B",
            markdown_description="Qwen first generation improved model with 110B parameters",
        ),
        cl.ChatProfile(
            name="Qwen1.5-72B",
            markdown_description="Qwen first generation improved model with 72B parameters",
        ),
        cl.ChatProfile(
            name="Qwen1.5-32B",
            markdown_description="Qwen first generation improved model with 32B parameters",
        ),
        cl.ChatProfile(
            name="Qwen1.5-2.7B",
            markdown_description="Qwen first generation improved model with 110B parameters",
        ),
        cl.ChatProfile(
            name="Qwen-72B",
            markdown_description="Qwen first generation model with 72B parameters",
        ),
        cl.ChatProfile(
            name="Qwen-14B",
            markdown_description="Qwen first generation model with 14B parameters",
        ),
        cl.ChatProfile(
            name="Qwen-7B",
            markdown_description="Qwen first generation model with 7B parameters",
        ),
        cl.ChatProfile(
            name="Llama-3.1-405B",
            markdown_description="Meta Open Source Model Llama with 405B parameters",
        ),
        cl.ChatProfile(
            name="Llama-3.1-70B",
            markdown_description="Meta Open Source Model Llama with 70B parameters",
        ),
        cl.ChatProfile(
            name="Llama-3.1-8B",
            markdown_description="Meta Open Source Model Llama with 8B parameters",
        ),
        cl.ChatProfile(
            name="Llama-3-70B",
            markdown_description="Meta Open Source model Llama-3 with 70B parameters",
        ),
        cl.ChatProfile(
            name='Aya-23B',
            markdown_description='Cohere open sourced AI model with 23B parameters'
        ),
        cl.ChatProfile(
            name="Llama-3-8B",
            markdown_description="Meta Open Source model Llama-2 with 7B parameters",
        ),
        cl.ChatProfile(
            name = "gemma2-9B",
            markdown_description = 'Google Generation 2 Open Source LLM with 9B parameters'
        ),
        cl.ChatProfile(
            name = "gemma-7B",
            markdown_description = 'Google Generation 1 Open Source LLM with 7B parameters'
        ),
        cl.ChatProfile(
            name="zephyr-7B",
            markdown_description="Open Source model Zephyr with 7B parameters",
        ),
        cl.ChatProfile(
            name='mistral-7B',
            markdown_description = 'mistral open source LLM with 7B parameters'
        ),
        # cl.ChatProfile(
        #     name="Toka-353M",
        #     markdown_description="PartAI Open Source model Toka with 353M parameters",
        # )
    ]

@cl.on_chat_start
async def on_chat_start():
    chat_profile = cl.user_session.get("chat_profile")
    if not chat_profile:
        await cl.Message(
            content='please choose a model to start'
        ).send()
    
    if chat_profile == 'neural-brain-AI':
        await cl.ChatSettings(
            [
            Select(
                id="NB-Model",
                label="NeuralBrain - Models",
                values=["Neural Brain AI"],
                initial_index=0,
                )
            ]
        ).send()
        await cl.Message(
            content="Hello, I am the main model of neural brain team, i am an instance of ChatGPT-4, This team finetuned me and i am ready to help you"
        ).send()

    if chat_profile == 'Dorna-AI':
        await cl.ChatSettings(
            [
             Select(
                    id="param_3",
                    label="Parameter 3",
                    values=["512"],  # Only one selectable value
                    initial_index=0,
                    tooltip="Config parameter 3 (e.g., max tokens)",
                ),
                Select(
                    id="param_4",
                    label="Parameter 4",
                    values=["0.7"],  # Only one selectable value
                    initial_index=0,
                    tooltip="Config parameter 4 (e.g., temperature)",
                ),
                Select(
                    id="param_5",
                    label="Parameter 5",
                    values=["0.95"],  # Only one selectable value
                    initial_index=0,
                    tooltip="Config parameter 5 (e.g., top_p)",
                ),
                Select(
                    id="api_name",
                    label="API Name",
                    values=["/chat"],
                    initial_index=0,
                ),
            ]
        ).send()

        await cl.Message(
            content='my name is Dorna, Your AI Assistant designed by neural nexus team. i was made by Artin Daneshvar and Sadra Noadoust, 2 iranian students!'
        ).send()
    if chat_profile == 'gpt4-o-mini':
        await cl.ChatSettings(
            [
                Select(
                    id="OpenAI-Model",
                    label="OpenAI - Model",
                    values=["gpt4-o-mini"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im one of the best models openai have released and i am configured by two iranian boys to help you."
        ).send()

    # if chat_profile == 'Image-Generation':
    #     image = cl.Image(path='cat.png', name="result", display="inline")
    #     await cl.Message(
    #         content="I can make high quality & resoloution images for you, This is an example of what i can do!",
    #         elements=[image],
    #         ).send()
    if chat_profile == 'GPT-4':
        await cl.ChatSettings(
            [
                Select(
                    id="OpenAI-Model",
                    label="OpenAI - Model",
                    values=["gpt-4"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im OpenAI's latest and biggest model.  i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()

    if chat_profile == 'gpt-3.5-turbo':
        await cl.ChatSettings(
            [
                Select(
                    id="OpenAI-Model",
                    label="OpenAI - Model",
                    values=["gpt-3.5-turbo"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im one of the OpenAI's models. one of the best models. i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
    # if chat_profile == 'GPT-3.5-turbo-0125':
    #     await cl.ChatSettings(
    #         [
    #             Select(
    #                 id="OpenAI-Model",
    #                 label="OpenAI - Model",
    #                 values=["gpt-3.5-turbo-0125"],
    #                 initial_index=0,
    #             ),
    #             Slider(
    #                 id="Temperature",
    #                 label="Model Temperature",
    #                 initial=0.7,
    #                 min=0,
    #                 max=1,
    #                 step=0.1,
    #             ),
    #         ]
    #     ).send()
        # await cl.Message(
        #     content="Im one of the OpenAI's models. one of the best models. i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        # ).send()

    if chat_profile == 'gpt-3.5-turbo-1106':
        await cl.ChatSettings(
            [
                Select(
                    id="OpenAI-Model",
                    label="OpenAI - Model",
                    values=["gpt-3.5-turbo-1106"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im one of the OpenAI's models. one of the best models. i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()

    # if chat_profile == 'davinci-002':
    #     await cl.ChatSettings(
    #         [
    #             Select(
    #                 id="OpenAI-Model",
    #                 label="OpenAI - Model",
    #                 values=["davinci-002"],
    #                 initial_index=0,
    #             ),
    #             Slider(
    #                 id="Temperature",
    #                 label="Model Temperature",
    #                 initial=0.7,
    #                 min=0,
    #                 max=1,
    #                 step=0.1,
    #             ),
    #         ]
    #     ).send()
    #     await cl.Message(
    #         content="Im one of the OpenAI's models. i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
    #     ).send()
    if chat_profile == 'TTS':
        await cl.Message(
            content="Im TTS. of the best models OpenAI ever created. i can convert text to speech! . i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()

    if chat_profile == 'Llama-3.1-405B':
        await cl.ChatSettings(
            [
                Select(
                    id="Meta-Model",
                    label="Meta - Model",
                    values=["Llama-3.1-405B"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im the big Llama-3.1!. one of the best open source models released by Meta! i am the Big version of meta's open source LLMs., i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
        
    if chat_profile == 'Llama-3.1-70B':
        await cl.ChatSettings(
            [
                Select(
                    id="Meta-Model",
                    label="Meta - Model",
                    values=["Llama-3.1-70B"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im the second-big Llama-3.1!. one of the best open source models released by Meta! i am the Big version of meta's open source LLMs., i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
    if chat_profile == 'Llama-3.1-8B':
        await cl.ChatSettings(
            [
                Select(
                    id="Meta-Model",
                    label="Meta - Model",
                    values=["Llama-3.1-8B"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im the small Llama-3.1!. one of the best open source models released by Meta! i am the Big version of meta's open source LLMs., i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
        
    if chat_profile == 'Llama-3-70B':
        await cl.ChatSettings(
            [
                Select(
                    id="Meta-Model",
                    label="Meta - Model",
                    values=["Llama-3-70B"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im the big Llama-3!. one of the best open source models released by Meta! i am the Big version of meta's open source LLMs., i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
    if chat_profile == 'Llama-3-8B':
        await cl.ChatSettings(
            [
                Select(
                    id="Meta-Model",
                    label="Meta - Model",
                    values=["Llama-3-8B"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()        
        await cl.Message(
            content="Im The small Llama!. one of the best open source models released by Meta! i am the small version of meta's open source LLMs. i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
    if chat_profile == 'gemma2-9B':
        await cl.ChatSettings(
            [
                Select(
                    id="Google-Model",
                    label="Google - Model",
                    values=["Gemma-9B"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im Gemma2. the 9B version of google second generation open source LLMs. i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
        
    if chat_profile == 'gemma-7B':
        await cl.ChatSettings(
            [
                Select(
                    id="Google-Model",
                    label="Google - Model",
                    values=["Gemma-7B"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im Gemma. the small version of google open source LLMs. i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
    if chat_profile == 'zephyr-7B':
        await cl.ChatSettings(
            [
                Select(
                    id="zephyr-Model",
                    label="zephyr - Model",
                    values=["zephyr-7B"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im Zephyr. One of the best open source LLMs. i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
    if chat_profile == 'mistral-7B':
        await cl.ChatSettings(
            [
                Select(
                    id="Mistral-Model",
                    label="Mistral - Model",
                    values=["Mistral-7B"],
                    initial_index=0,
                ),
                Slider(
                    id="Temperature",
                    label="Model Temperature",
                    initial=0.7,
                    min=0,
                    max=1,
                    step=0.1,
                ),
            ]
        ).send()
        await cl.Message(
            content="Im Mistral. the small version of Mistral Family. i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? "
        ).send()
    # if chat_profile == 'Toka-353M':
    #     await cl.ChatSettings(
    #         [
    #             Select(
    #                 id="PartAI-Model",
    #                 label="PartAI - Model",
    #                 values=["TokaBert-353M"],
    #                 initial_index=0,
    #             ),
    #             Slider(
    #                 id="Temperature",
    #                 label="Model Temperature",
    #                 initial=0.7,
    #                 min=0,
    #                 max=1,
    #                 step=0.1,
    #             ),
    #         ]
    #     ).send()
    #     await cl.Message(
    #         content="Im Toka. An opens source  persian LLM . i was configured by Artin Daneshvar and Sadra Noadoust, 2 iranian students to help you, how can i assist you today ? you should ask me your questions like : the capital of england is <mask> "
    #     ).send()

@cl.on_message  
async def main(message: cl.Message):
    chat_profile = cl.user_session.get("chat_profile")
    if not chat_profile or chat_profile == 'None':
        await cl.Message(
            content="Please select a model first."
        ).send()
        return
    if chat_profile == 'neural-brain-AI':
        completion = openai_client.chat.completions.create(
            model="ft:gpt-3.5-turbo-1106:nb:aria1:9UWDrLJK",
            messages=[
                {"role": "system", "content": "You are neural nexus official chatbot, you are made by Artin Daneshvar and Sadra Noadoust"},
                {"role": "user", "content": message.content}
            ]
        )
        model_response = completion.choices[0].message.content
        await cl.Message(
            content=model_response
        ).send()

    elif chat_profile == "Dorna-AI":
        result = hf_text_client.predict(
            message=message.content,
            request="your name is Dorna,An AI Assistant designed by neural nexus team. i was made by Artin Daneshvar and Sadra Noadoust, 2 iranian students!",
            param_3=512,
            param_4=0.7,
            param_5=0.95,
            api_name="/chat"
        )
        model_response = result.strip("</s>") 
        await cl.Message(
            content=model_response
        ).send()
    elif chat_profile == "gpt4-o-mini":
        completion = openai_client.chat.completions.create(
            model="gpt-4o-mini",
            messages=[
                {"role": "system", "content": "You are neural nexus official chatbot, you are made by Artin Daneshvar and Sadra Noadoust"},
                {"role": "user", "content": message.content}
            ]
        )
        model_response = completion.choices[0].message.content
        await cl.Message(
            content=model_response
        ).send()
  #   elif chat_profile == 'Image-Generation':
  #       result = hf_image_client.predict(
		# prompt=message.content,
		# negative_prompt="",
		# seed=0,
		# randomize_seed=True,
		# width=512,
		# height=512,
		# guidance_scale=0,
		# num_inference_steps=2,
		# api_name="/infer"
  #       )
  #       image = cl.Image(path=result, name="result", display="inline")
  #       await cl.Message(
  #           content="This message has an image!",
  #           elements=[image],
  #       ).send()
    elif chat_profile == 'GPT-4':
        completion = openai_client.chat.completions.create(
            model="gpt-4",
            messages=[
                {"role": "system", "content": "You are neural nexus official chatbot, you are made by Artin Daneshvar and Sadra Noadoust"},
                {"role": "user", "content": message.content}
            ]
        )
        model_response = completion.choices[0].message.content
        await cl.Message(
            content=model_response
        ).send()

    elif chat_profile == 'gpt-3.5-turbo':
        completion = openai_client.chat.completions.create(
            model="gpt-3.5-turbo",
            messages=[
                {"role": "system", "content": "You are neural nexus official chatbot, you are made by Artin Daneshvar and Sadra Noadoust"},
                {"role": "user", "content": message.content}
            ]
        )
        model_response = completion.choices[0].message.content
        await cl.Message(
            content=model_response
        ).send()
    elif chat_profile == 'GPT-3.5-turbo-0125':
        completion = openai_client.chat.completions.create(
            model="GPT-3.5-turbo-0125",
            messages=[
                {"role": "system", "content": "You are neural nexus official chatbot, you are made by Artin Daneshvar and Sadra Noadoust"},
                {"role": "user", "content": message.content}
            ]
        )
        model_response = completion.choices[0].message.content
        await cl.Message(
            content=model_response
        ).send()
    elif chat_profile == 'gpt-3.5-turbo-1106':
        completion = openai_client.chat.completions.create(
            model="gpt-3.5-turbo-1106",
            messages=[
                {"role": "system", "content": "You are neural nexus official chatbot, you are made by Artin Daneshvar and Sadra Noadoust"},
                {"role": "user", "content": message.content}
            ]
        )
        model_response = completion.choices[0].message.content
        await cl.Message(
            content=model_response
        ).send()
    # elif chat_profile == 'davinci-002':
    #     completion = openai_client.chat.completions.create(
    #         model="davinci-002",
    #         messages=[
    #             {"role": "system", "content": "You are neural nexus official chatbot, you are made by Artin Daneshvar and Sadra Noadoust"},
    #             {"role": "user", "content": message.content}
    #         ]
    #     )
    #     model_response = completion.choices[0].message.content
    #     await cl.Message(
    #         content=model_response
    #     ).send()

    elif chat_profile == 'TTS':
        response = openai_client.audio.speech.create(
            model="tts-1",
            voice="alloy",
            input=message.content,
        )

        response.stream_to_file("output.mp3")

        elements = [
            cl.Audio(name="output.mp3", path="./output.mp3", display="inline"),
        ]
        await cl.Message(
            content="Here it is the response!",
            elements=elements,
        ).send()

    elif chat_profile == 'Llama-3.1-405B':        
        client = InferenceClient(
            "meta-llama/Meta-Llama-3.1-405B-Instruct",
            token=f'{hf_token_llama_3_1}',
        )
        
        for message in client.chat_completion(
            messages=[{"role": "user", "content": f'{message.content}'}],
            max_tokens=500,
            stream=True,
        ):
            complete_message += message.choiches[0].delta.content
            await cl.Message(
                content=complete_message,
            ).send()

    elif chat_profile == 'Llama-3.1-70B':
        completion = groq_client.chat.completions.create(
            model="llama-3.1-70b-versatile",
            messages=[
                {
                    "role": "user",
                    "content": message.content
                }
            ],
            temperature=1,
            max_tokens=1024,
            top_p=1,
            stream=True,
            stop=None,
        )
    
        complete_content = ""
    
        # Iterate over each chunk
        for chunk in completion:
            # Retrieve the content from the current chunk
            content = chunk.choices[0].delta.content
            
            # Check if the content is not None before concatenating it
            if content is not None:
                complete_content += content
    
        # Send the concatenated content as a message
        await cl.Message(content=complete_content).send()

    elif chat_profile == 'Llama-3.1-8B':
        completion = groq_client.chat.completions.create(
            model="llama-3.1-8b-instant",
            messages=[
                {
                    "role": "user",
                    "content": message.content
                }
            ],
            temperature=1,
            max_tokens=1024,
            top_p=1,
            stream=True,
            stop=None,
        )
    
        complete_content = ""
    
        # Iterate over each chunk
        for chunk in completion:
            # Retrieve the content from the current chunk
            content = chunk.choices[0].delta.content
            
            # Check if the content is not None before concatenating it
            if content is not None:
                complete_content += content
    
        # Send the concatenated content as a message
        await cl.Message(content=complete_content).send()

    elif chat_profile == 'Llama-3-70B':
        completion = groq_client.chat.completions.create(
            model="llama3-70b-8192",
            messages=[
                {
                    "role": "user",
                    "content": message.content
                }
            ],
            temperature=1,
            max_tokens=1024,
            top_p=1,
            stream=True,
            stop=None,
        )

        complete_content = ""

        # Iterate over each chunk
        for chunk in completion:
            # Retrieve the content from the current chunk
            content = chunk.choices[0].delta.content
            
            # Check if the content is not None before concatenating it
            if content is not None:
                complete_content += content

        # Send the concatenated content as a message
        await cl.Message(content=complete_content).send()

    elif chat_profile == 'Llama-3-8B':
        completion = groq_client.chat.completions.create(
            model="llama3-8b-8192",
            messages=[
                {
                    "role": "user",
                    "content": message.content
                }
            ],
            temperature=1,
            max_tokens=1024,
            top_p=1,
            stream=True,
            stop=None,
        )

        complete_content = ""

        # Iterate over each chunk
        for chunk in completion:
            # Retrieve the content from the current chunk
            content = chunk.choices[0].delta.content
            
            # Check if the content is not None before concatenating it
            if content is not None:
                complete_content += content

        # Send the concatenated content as a message
        await cl.Message(content=complete_content).send()

    elif chat_profile == 'gemma2-9B':
        completion = groq_client.chat.completions.create(
            model="gemma-9b-it",
            messages=[
                {
                    "role": "user",
                    "content": message.content
                }
            ],
            temperature=1,
            max_tokens=1024,
            top_p=1,
            stream=True,
            stop=None,
        )

        complete_content = ""

        # Iterate over each chunk
        for chunk in completion:
            # Retrieve the content from the current chunk
            content = chunk.choices[0].delta.content
            
            # Check if the content is not None before concatenating it
            if content is not None:
                complete_content += content

        # Send the concatenated content as a message
        await cl.Message(content=complete_content).send()

    elif chat_profile == 'gemma-7B':
        completion = groq_client.chat.completions.create(
            model="gemma-7b-it",
            messages=[
                {
                    "role": "user",
                    "content": message.content
                }
            ],
            temperature=1,
            max_tokens=1024,
            top_p=1,
            stream=True,
            stop=None,
        )

        complete_content = ""

        # Iterate over each chunk
        for chunk in completion:
            # Retrieve the content from the current chunk
            content = chunk.choices[0].delta.content
            
            # Check if the content is not None before concatenating it
            if content is not None:
                complete_content += content

        # Send the concatenated content as a message
        await cl.Message(content=complete_content).send()
    
    elif chat_profile == "zephyr-7B":
        result = hf_text_client.predict(
            message=message.content,
            request="your name is zephyr,An AI Assistant designed by neural nexus team. i was made by Artin Daneshvar and Sadra Noadoust, 2 iranian students!",
            param_3=512,
            param_4=0.7,
            param_5=0.95,
            api_name="/chat"
        )
        model_response = result.strip("</s>") 
        await cl.Message(
            content=model_response
        ).send()

    elif chat_profile == 'mistral-7B':
        completion = groq_client.chat.completions.create(
            model="mixtral-8x7b-32768",
            messages=[
                {
                    "role": "user",
                    "content": message.content
                }
            ],
            temperature=1,
            max_tokens=1024,
            top_p=1,
            stream=True,
            stop=None,
        )

        complete_content = ""

        for chunk in completion:
            content = chunk.choices[0].delta.content
            
            if content is not None:
                complete_content += content

        await cl.Message(content=complete_content).send() 

    # elif chat_profile == 'Toka-353M':
    #     output = query({
    #     "inputs": message.content,
    # })
    #     await cl.Message(
    #         content=output[0]['sequence']
    #     ).send()

    elif chat_profile == 'Aya-23B':
        stream = co.chat_stream( 
        model='c4ai-aya-23',
        message=message.content,
        temperature=0.3,
        # chat_history=[{"role": "User", "message": "Hello"}, {"role": "Chatbot", "message": "Hello! How can I help you today?"}, {"role": "User", "message": "Hi"}, {"role": "User", "message": "hello"}],
        prompt_truncation='OFF',
        connectors=[],
        ) 

        complete_content = ''
        for event in stream:
            if event.event_type == 'text-generation':
                complete_content += event.text
        await cl.Message(content=complete_content).send()



@cl.on_settings_update
async def setup_agent(settings):
    print("on_settings_update", settings)