LM Studio is a tool that allows you to run local Large Language Models (LLMs) on your computer. One of the best things about this utility is that it allows you to run the models locally to ensure maximum privacy. In this guide, we are going to configure LM Studio to use multiple AI’s on an offline PC, so, if you are interested, continue reading.
Configure LM Studio to use multiple AI’s on offline PC
LM Studio allows you to download and run large language models on your computer without needing the internet. It helps keep your data private by processing everything locally. With it, you can use multiple AI models for tasks like content creation and data analysis. If you want to configure LM Studio to use multiple AI’s on offline PC, follow the steps mentioned below.
- Install LM Studio
- Install multiple models
- Start using the model
Let us discuss them in detail.
1] Install LM Studio
First of all, we need to get the LM Studio tool from their official website. So, for that, you need to go lmstudio.ai. Since I am using Windows, I will select Download LM Studio for Windows, but depending on your platform, you need to download the LM Studio version. Once done, this will trigger the download of the respected exe file.
After downloading the LM Studio installation media, double-click on it, and then select the user for which you want to install the application. If you want to install it for all the users, you need admin access as it will get a UAC prompt. Since I have access, I did that, then selected the folder, and finally launched the application.
2] Install multiple models
Once you launch the LM Studio, you will be prompted to download your first LLM Models, click on Get your first LLM, and then download the suggested model. In case, you want to select your own model, click on Skip on boarding.
To download any LLM model, you need to click on the Discover (magnifying glass) icon. Now, search for the model you want to download. One of the best things about the LM Studio is the fact that it checks your system details and tell you if the model is too large for your machine. For example, the DeepSeek Coder V2 Lite is too large for my device, whereas, Llama 3.2 1B will run smoothly.
Once you know which model will work on your machine, click on the Download icon and let it download the model. Depending on your network bandwidth, it may take some time.
This will download the LLM Model along with the Runtime. However, if you want to upgrade to a different model, in my case, let’s say, I wanted to install CPU llama.cpp as I don’t have a powerful GPU and GPU-Acceleration isn’t something that my machine can handle, I went to the Runtime tab, and clicked on Update button next to Runtime model.
In order to download another model, you need to go to the Discover button again, look for the model you want to download. Then, click on the Download button associated with it. Wait for a few minutes as it downloads your model. Once it is downloaded, you can go to the Runtime download any updates if available, and if needed, and then, all you have to do is click on Load Models. This will start the server of that model locally on your computer.
Note: If you closed the Discover prompt without starting the model, go to the Download icon from the bottom left corner of the screen, and then select Load Model associated with your model.
3] Start using the model
After downloading the models, all you need to do is go to the Chat icon. Then, look at the top of the window, and you will see all your models, click on the Model that you want to select.
You can even change the preferences of the model before using them. For that, click on the side arrow that is next to the model and then change settings such as GPU Offload, CPU Thread Pool Size, Context Length, and other options. Once done, just click on Load Model.
Once the model is loaded, go to the Chatbot and start chatting with the AI Model. Make sure to note the RAM and CPU count at the bottom right corner of the window.
Hopefully, with the help of this guide, you will be able to run multiple LLM Models in LM Studio. However, there are other tools to run LLM locally, you should check them out as well.
Read: Is MSTY LLM for Windows safe? How to install it?
Does LM Studio work offline?
Yes, LM Studio works offline. Once you’ve downloaded the necessary models and runtimes, you can use it entirely without an internet connection. All processing is done locally on your machine, ensuring that your data remains private.
Read: Create AI agents using LLMs Claude, Copilot, ChatGPT, Gemini
How much RAM do I need for LM studio?
Every model in LM Studio has different RAM requirements. However, we recommend you have at least 16 GB of RAM if you want to run any model without issues. Additionally, if you have the capacity to add a GPU, do that, too.
Also Read: Run Qwen AI locally on Windows 11.
