Using Chatbots

Links to Selected Chatbots

Claude

From Anthropic, Claude comes as several LLMs. There is a free version, and Claude Pro costs $18/month.

• Claude 3.5 Haiku
• Claude 3.5 Opus (premium)
• Claude 3.5 Sonnet (premium)

CoPilot

Microsoft's derivative of OpenAI's ChatGPT is being integrated into many products, and its capabilities expanded to include vision, image generation, etc.

Gemini

From Google, Gemini comes in various flavors. Probably the best way to learn about them is in Google's AI Studio. An API is available to download from this site.

Grok

Accessed via the sidebar on the X platform, Grok is free to use, and early versions can be downloaded for custom use on one's own hardware.

Llama

An open-source LLM from Meta, using it is free.

Leo

Leo is the AI agent built into Brave broswer. It can use several different LLMs:

• Mixtral
• Claude 3 Haiku
• Claude 3.5 Sonnet (premium)
• Llama 3.1 8B

Basic use is free, but Claude 3.5 Sonnet requires a premium subscription ($149/year or $14.99/month).\

MetaAI

Another open-source LLM from Meta, MetaAI is accessible via Facebook.

Mistral AI

A French AI with several open-source LLMs that are available via Leo AI and OnlyOffice AI plugin, among others

OpenAI

The home of ChatGPT, this may be the best-known AI chatbot currently thanks to it's first-out-of-the-gate status.

Perplexity

Said to be currently the best for research because it searches the internet and provides citations, Perplexity Pro costs $16/seat/month, although there is also a free option.

Using Ollama, download a library to use in the Ollama Docker image, or just use the built-in default (llama3).

Pi

A free AI that is described as "your personal AI".

Ollama

Ollama is an open-source CLI app that implements chatbots using various LLMs without requiring APIs. There is an official Ollama Docker image available that has an extensive library of LLMs available on GitHub. Please note: Docker must first be installed to implement the Docker image.

Note about Docker

Docker is an open-source, cross-platform application for creating and using containers. Installing and using Docker is beyond the scope of this website, but there are many resources available online and in books for learning how to implement this very useful software tool.

Running Ollama in a Docker Container

These instructions are for Linux. To use the Ollama Docker image on Windows or MacOS, visit the Ollama GitHub for instructions.

To use Ollama in Docker, use these Terminal commands:

• Pull a library: docker exec -it ollama ollama pull [name of LLM]
• Run the model: docker exec -it ollama ollama run [name of LLM]

Once a model has been pulled and run, the following commands are available:

Available Commands:
  /set            Set session variables
  /show           Show model information
  /load <model>   Load a session or model
  /save <model>   Save your current session
  /clear          Clear session context
  /bye            Exit
  /?, /help       Help for a command
  /? shortcuts    Help for keyboard shortcuts

---

Back to AI & Machine Learning