A Virtual Chat assistant you can ask unlimited questions to. If you're not aware of what ollama or llama 3, check out they're GitHub for details on how it works. I just slapped it into a Discord bot. Quite simply, they're chatgpt but open source, so its free unlimited usage for developers and you! No OpenAI Token required.
- Install Docker
- Create an Ollama container.
docker run -d -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama
- Click the Code button.
- Click Download ZIP.
- Navigate to the downloaded zip file.
- Unzip the file.
- Navigate into the llama3-bot folder until you see package.json
- Open quickstart.bat
The options within the config file you're free to change however you'd like, just note the prefix must be lowercase as suggested by its name.
The Program string will tell the AI how to act, if you want it to rude, you would tell it there. I chose to tell it what it was instead of asking it to act a certain way, but I'm not certain it makes a difference.
The prefix is, simply how you access it. I really hate prefixes, especially for FAQ/Chat bots, feels to clunky, I prefer a more alexa/google home/smart assistant feel, but its up to you how it works. It checks the start of the users message for the "prefix" so ensure you don't use a phrase/letter/word that you know is used frequently in your servers.
Simple config for a simple bot.
Open an issue if you have an issue or a feature request. Make a fork and open a pull request if you have a change.
I likely won't be doing too much feature changes on my own, I'll accept pull requests if they're valuable, and I'll look into feature requests, but I don't think there is much more to change. I'll likely only do package upgrading, etc.