https://github.com/soodaayush/llamatalk

A chatbot powered by Meta's Codellama.
https://github.com/soodaayush/llamatalk

codellama css html javscript ollama scss

Last synced: about 1 year ago
JSON representation

A chatbot powered by Meta's Codellama.

• Host: GitHub
• URL: https://github.com/soodaayush/llamatalk
• Owner: soodaayush
• Created: 2024-02-01T21:39:50.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2025-03-08T23:57:09.000Z (about 1 year ago)
• Last Synced: 2025-03-09T00:24:28.674Z (about 1 year ago)
• Topics: codellama, css, html, javscript, ollama, scss
• Language: SCSS
• Homepage:
• Size: 66.4 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# LlamaTalk

![llamatalk](https://github.com/user-attachments/assets/d2fdb3f1-6871-4da9-8c0c-b210b9ba8be3)

## Inspiration

I wanted to create my own local chatbot with an interface. When I discovered that it was not difficult due to an easy-to-use service called Ollama, I decided to experiment with making a local chatbot website.

## Challenges

Connecting the Large Language Model (LLM) to the website itself and ensuring the correct objects were used within the response object.

## Lessons Learned

I learned about using the Ollama service and how to send and fetch data from the LLM.

## The Website

A chatbot powered by Meta's Codellama. You can ask it a question, and you will get a response. As of now, you must run the LLM locally on your machine through a service called Ollama. From there, your LLM will be ran on a localhost server, which you can define in the JavaScript file. LlamaTalk uses the localhost to provide a prompt and receive a response. You can customize the code to use any model with any parameter count as you please.

## Links

Ollama: https://ollama.com/