The AI circle has been abuzz again in recent days because Meta (formerly Facebook) has open sourced Llama 3 8B and 70B models of two different sizes. According to Meta, this model was trained using 24,576 GPUs.
The focus is on open source, which is really awesome!
So as ordinary people, can we use Llama 3 on our own computers? This is definitely possible, and you don’t need to know programming. You can run it locally by just clicking around with the mouse, and it can be used even when the network is disconnected.
It’s super easy to use Llama 3 on your own computer, just 2 steps.
Install GPT4All client.
Download the Llama 3 model.
Step 1: Download GPT4All
Download the GPT4All client, the address is here: gpt4all.io/index.html
Choose the installation package according to your system. After the download is complete, open the installation package and click “Next” all the way. It’s very simple ~
Step 2: Download the Llama 3 model
After installing GPT4All, open it.
When you open it for the first time, because you have not downloaded any models, this pop-up window will pop up to let you choose one to download.
Select the first Llama 3 Instruct here, then click the Download button on the right and wait for it to download before you can use it.
Once the model is downloaded, select it at the top of the interface and start chatting.
