Use “Yang” to Communicate With AI Without Worrying About Privacy
Like the idea of artificial intelligence, but would you like to use it without trusting some large server? You can run large language models locally, getting something like ChatGPT that runs completely offline.
Jan is a free and open source application that makes it easy to download several large language models and start communicating with them. There are simple installers for Windows, macOS and Linux. It’s not ideal. The models are not necessarily as good as the latest OpenAI or Google models, and depending on the power of your computer, it may take a while to get results. On the other hand, you can use this technology without having to worry about the privacy and security issues associated with using an online AI service.
After installing Ian you’ll need to choose a model – if you don’t know which one, I’d start with the Mistral at the top of the list – you can always try something else later if you don’t like the result. . Once the model is loaded, you can start communicating.
If you want, you can provide some general instructions for the bot in the right panel – the default is set to “you are a helpful helper” but you can change it to whatever you like to give a little more context. After this, you can start using the service just like ChatGPT or Google Gemini. I tried to ask for a summary of my recent article ; it did a decent job.
Again, if you don’t like the results or think it takes too long to get results, try a few different models. All of them are free and optimized for different purposes: some of them are specifically designed for coding issues, for example, and some are optimized to run on computers without a lot of CPU power. It’s a matter of finding what works best for you.
And there is another interesting feature: if the application is running, it can also act as the equivalent OpenAI API. If you don’t know what this means, don’t worry – it’s a very strange thing. In short, however, this means you can use Jan in apps that would otherwise require a paid ChatGPT subscription – just enable the API feature in Settings and point other apps to your local IP address and port number instead of OpenAI.