Articles → LANGCHAIN → Create A Blog Generator Using Streamlit, LLMA, And Langchain

Create A Blog Generator Using Streamlit, LLMA, And Langchain

This article explains how to create a blog generator using Streamlit, LLMA, and Langchain.

Powershell Commands

To create the infrastructure, we will use Microsoft Azure. To begin, open Windows PowerShell in Admin mode and connect to Azure.

Create A Resource Group

Execute the following command in PowerShell.

New-AzResourceGroup -Name "streamlit-rg" -Location "CentralIndia"

Create An Ubuntu Virtual Machine

To create a virtual machine, execute the following command: -

New-AzVM `
    -ResourceGroupName "streamlit-rg" `
    -Name "streamlit-vm" `
    -Location "CentralIndia" `
    -Image "Ubuntu2204" `
    -Size "Standard_D4s_v3" `
    -PublicIpAddressName "streamlit-ip" `
    -VirtualNetworkName "streamlit-vnet" `
    -SubnetName "default" `
    -SecurityGroupName "streamlit-nsg" `
    -OpenPorts 22,8501 `
    -Credential (Get-Credential)

When the above command is executed, a prompt will appear to enter the username and password. This username and password will be used to login to the Ubuntu virtual machine.

Please note that ports 22 and 8501 were configured to be open when the virtual machine was created.

Create A Public IP Address

To create a public IP address, execute the following command in PowerShell.

Get-AzPublicIpAddress -ResourceGroupName "streamlit-rg" -Name "streamlit-ip" | Select IpAddress

Once you execute the above command, the public IP address is created and displayed on the PowerShell console.

Create Folders

To create a folder, login to the virtual machine using the following command: -

ssh [usermame]@[public_ip_address]

So, if the username is azureuser and public IP address is 20.219.62.72. Then the command would be: -

ssh azureuser@20.219.62.72

Once the above command is executed, the console will ask for the password. Enter the password to login.

Create the streamlit-app folder using the following command: -

mkdir streamlit-app

Once the folder is created, go to the folder and inside it and create another folder.

cd streamlit-app
mkdir models

Download Model

Once you are inside the models folder execute the following command to download the model.

wget https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGUF/resolve/main/llama-2-7b-chat.Q4_K_M.gguf

Create App.Py Inside Streamlit-App Folder

To create the file, write the following command: -

nano app.py

A screen will appear to enter the content. Write the following code.

import streamlit as st
from langchain_core.prompts import PromptTemplate
from langchain_community.llms import CTransformers

# ---------------- Load Model ----------------

def load_model():
    return CTransformers(
    model="/home/azureuser/streamlit-app/models/llama-2-7b-chat.Q4_K_M.gguf",
    model_type="llama",
    config={
        "max_new_tokens": 256,
        "temperature": 0.01
    }
)

# ---------------- LLaMA/Mistral Response Function ----------------

def getLLamaresponse(input_text, no_words, blog_style):
    llm = load_model()

    template = (
        "Write a blog for {blog_style} job profile "
        "for a topic \"{input_text}\" within {no_words} words."
    )

    prompt = PromptTemplate(
        input_variables=["blog_style", "input_text", "no_words"],
        template=template
    )

    response = llm.invoke(
        prompt.format(
            blog_style=blog_style,
            input_text=input_text,
            no_words=no_words
        )
    )

    return response

# ---------------- Streamlit UI ----------------

st.set_page_config(
    page_title="Generate Blogs",
    page_icon="📝",
    layout="centered",
    initial_sidebar_state="collapsed"
)

st.header("Generate Blogs")

input_text = st.text_input("Enter the Blog Topic")

col1, col2 = st.columns(2)

with col1:
    no_words = st.text_input("No of Words", value="300")

with col2:
    blog_style = st.selectbox(
        "Writing the blog for",
        ("Researchers", "Data Scientist", "Common People"),
        index=0
    )

if st.button("Generate"):
    if not input_text.strip():
        st.error("Please enter a blog topic.")
    elif not no_words.isdigit():
        st.error("Please enter a valid number of words.")
    else:
        with st.spinner("Generating blog..."):
            response = getLLamaresponse(
                input_text,
                no_words,
                blog_style
            )

        st.success("Blog generated successfully!")
        st.write(response)

Press Ctrl + O to save the file. Press Enter to confirm the filename. Press Ctrl + X to exit the Nano editor.

Create Requirements.Txt Inside Streamlit-App Folder

Similarly, create requirements.txt with the following content.

streamlit
langchain-core
langchain-community
ctransformers

Execute Commands Inside The Streamlit-App Folder

Execute the following commands.

sudo apt update
sudo apt install -y python3 python3-pip
pip install -r requirements.txt

Run The Application

Execute the following command to run the application.

streamlit run app.py --server.port 8501 --server.address 0.0.0.0

If you are getting a message - streamlit: command not found

echo 'export PATH=$PATH:$HOME/.local/bin' >> ~/.bashrc

source ~/.bashrc

Output

Picture showing the blog generation screen when page is loaded for the first time

Picture showing the screen after blog generation

Posted By -	Karan Gupta

Posted On -	Monday, August 3, 2026

Query/Feedback

Your Email Id		**

Subject		*

Query/Feedback	Characters remaining 250	**