Build Your First Streamlit LLM App

Hands-On with LLMs: A Project-Based Approach to Building LLM Applications

Explore Culinary Convo on GitHub

As a big fan of project-based learning, I believe that hands-on experience is the key to truly understanding complex concepts. That's why I built Culinary Convo, a multi-page Streamlit application to dive deep into the world of Large Language Models (LLMs), OpenAI's Chat completions API, PromptLayer, LangChain, and Streamlit's interactive web framework.

Streamlit: A Framework for Web Apps

Streamlit empowers developers to create web applications with ease. Its interactivity and simplicity were vital in building Culinary Convo, enabling users to experience the power of LLMs firsthand.

OpenAI's Chat Completions API

Chat models transform a list of messages into model-generated text, a versatile approach for both multi-turn and single-turn tasks.

PromptLayer

This tool records all your OpenAI API requests, allowing you to search and explore request history - a powerful asset for prompt management!

LangChain

LangChain provides the tools to build applications powered by LLMs. It's an orchestration tool that enables interactive chaining of different prompts.

Culinary Convo: A Technical Overview

A multi-page Streamlit application that brings together diverse technologies, including:

Features

• Generate Ingredient Replacements: Spice up your recipes with creative twists!
• Generate Themed Blog Posts for Your Recipes: Share your culinary stories!
• Version Control Your Prompts: Keep track of your creative expressions!
• Scrape Your Favorite Recipes: Your favorite dishes at your fingertips!
• Send Grocery Lists to Whatsapp: Make shopping a breeze!

Demo

Get a glimpse of Ingredients Guru and Gourmet Ghostwriter in action!

Step by Step: How to Run Culinary Convo

1. API Keys: Get free keys from OpenAI & PromptLayer.
2. Environment Variables: Add keys to .env file or .streamlit/secrets.toml.
3. Clone the Repo:git clone https://github.com/pranavprajapati/culinary_convo.git
4. Install Packages:cd repository conda create --name convo conda activate convo conda install --file requirements.txt
5. Start the Application:streamlit run Home.py

Roadmap

Future updates include support for different LLM providers, custom chatbots, and more.

Join the Journey

Explore Culinary Convo on GitHub and embark on a journey where technology meets creativity. Whether you're looking to build your first LLM app or explore prompt engineering, this project is a fascinating starting point.

Please create a pull request if you wish to contribute!

#LLMs #OpenAI #PromptLayer #LangChain #Streamlit #Tutorial #Innovation