How to download and run local LLMs.
One big question that always comes up on the topic of AI is are large language models able to run if they;re not connceted to the internet? The answer is yes and in this tutorial we'll show you how you can do it using LM Studio.
Step 1: Visit LM Studio
- Go to LM Studio on your web browser.
- Choose your platform (Mac, Windows, or Linux) and perform a standard install.
Step 2: Open LM Studio Application
- Once installed, open LM Studio.
- This will bring you to the main page of the application.
Step 3: Explore and Select Models
- Search for models using keywords or paste a Hugging Face repository URL.
- Browse "New and Noteworthy" for popular models voted by the community, like Open Hermes, Mistral, and Code Llama 2.
Step 4: Download a Model
- Choose a model to download, for example, "Code Llama."
- Click "Download." The model will appear in your downloads section.
Step 5: Using the Downloaded Model
- After downloading, navigate to the left-hand side and click on the "New Chat" speech bubble icon.
- For first-time users, a blank screen appears.
- Select the model you wish to use, such as OpenAI's GPT-3.5.
Step 6: Starting a Conversation
- Enter a prompt in the provided text box. For example: "Tell me about Earth."
- The model will generate a response based on your prompt.
Step 7: Understanding the Local Setup
- Note that these models run locally on your laptop or PC.
- Local versions may be slower compared to hosted versions like GPT-3.5 Turbo.
- No internet connection is required for local models.
Step 8: Additional Features
- Explore further settings and details on the left panel.
- Manage and set defaults for your downloaded models.
- Search and sort models by popularity, recent updates, or downloads.
Step 9: Interacting with the Response
- Once a response is generated, you can regenerate it, continue the chat, or tweak the prompts.
- Adjust model initialization, hardware settings, or save presets.
Step 1: Visit LM Studio
- Go to LM Studio on your web browser.
- Choose your platform (Mac, Windows, or Linux) and perform a standard install.
Step 2: Open LM Studio Application
- Once installed, open LM Studio.
- This will bring you to the main page of the application.
Step 3: Explore and Select Models
- Search for models using keywords or paste a Hugging Face repository URL.
- Browse "New and Noteworthy" for popular models voted by the community, like Open Hermes, Mistral, and Code Llama 2.
Step 4: Download a Model
- Choose a model to download, for example, "Code Llama."
- Click "Download." The model will appear in your downloads section.
Step 5: Using the Downloaded Model
- After downloading, navigate to the left-hand side and click on the "New Chat" speech bubble icon.
- For first-time users, a blank screen appears.
- Select the model you wish to use, such as OpenAI's GPT-3.5.
Step 6: Starting a Conversation
- Enter a prompt in the provided text box. For example: "Tell me about Earth."
- The model will generate a response based on your prompt.
Step 7: Understanding the Local Setup
- Note that these models run locally on your laptop or PC.
- Local versions may be slower compared to hosted versions like GPT-3.5 Turbo.
- No internet connection is required for local models.
Step 8: Additional Features
- Explore further settings and details on the left panel.
- Manage and set defaults for your downloaded models.
- Search and sort models by popularity, recent updates, or downloads.
Step 9: Interacting with the Response
- Once a response is generated, you can regenerate it, continue the chat, or tweak the prompts.
- Adjust model initialization, hardware settings, or save presets.
Step 1: Visit LM Studio
- Go to LM Studio on your web browser.
- Choose your platform (Mac, Windows, or Linux) and perform a standard install.
Step 2: Open LM Studio Application
- Once installed, open LM Studio.
- This will bring you to the main page of the application.
Step 3: Explore and Select Models
- Search for models using keywords or paste a Hugging Face repository URL.
- Browse "New and Noteworthy" for popular models voted by the community, like Open Hermes, Mistral, and Code Llama 2.
Step 4: Download a Model
- Choose a model to download, for example, "Code Llama."
- Click "Download." The model will appear in your downloads section.
Step 5: Using the Downloaded Model
- After downloading, navigate to the left-hand side and click on the "New Chat" speech bubble icon.
- For first-time users, a blank screen appears.
- Select the model you wish to use, such as OpenAI's GPT-3.5.
Step 6: Starting a Conversation
- Enter a prompt in the provided text box. For example: "Tell me about Earth."
- The model will generate a response based on your prompt.
Step 7: Understanding the Local Setup
- Note that these models run locally on your laptop or PC.
- Local versions may be slower compared to hosted versions like GPT-3.5 Turbo.
- No internet connection is required for local models.
Step 8: Additional Features
- Explore further settings and details on the left panel.
- Manage and set defaults for your downloaded models.
- Search and sort models by popularity, recent updates, or downloads.
Step 9: Interacting with the Response
- Once a response is generated, you can regenerate it, continue the chat, or tweak the prompts.
- Adjust model initialization, hardware settings, or save presets.