> ## Documentation Index
> Fetch the complete documentation index at: https://docs.alterhq.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Keep Your Data Private with Local AI

> How to run AI models on your own Mac for sensitive documents and confidential work

<Info>
  **Perfect for:** Lawyers, doctors, security researchers, or anyone working with sensitive data who doesn't want it leaving their computer.
</Info>

## The Concern

You're working with:

* Client confidential information
* Medical records
* Proprietary business data
* Personal financial documents
* Security research

And you're worried about sending this data to cloud AI services like OpenAI or Google, even if they claim it's private.

**The good news:** You don't have to. You can run AI models directly on your Mac.

## How Local Models Work

Instead of sending your data to the internet, you download an AI model to your Mac (like downloading a large app) and run it locally. Your data never leaves your computer.

<CardGroup cols={2}>
  <Card title="100% Private" icon="shield-check">
    Nothing sent to the cloud. Your data stays on your Mac.
  </Card>

  <Card title="Works Offline" icon="wifi-slash">
    Use AI on airplanes, in secure facilities, or anywhere without internet.
  </Card>

  <Card title="No API Costs" icon="money-bill">
    Once downloaded, no per-use fees. Analyze thousands of documents for free.
  </Card>

  <Card title="Faster for Big Files" icon="bolt">
    No upload/download time. Great for large documents.
  </Card>
</CardGroup>

## Setup: Two easy options

### Option 1: Ollama (Recommended for beginners)

Ollama makes running local models as simple as installing an app.

<Steps>
  <Step title="Install Ollama">
    1. Download from [ollama.com](https://ollama.com)
    2. Install it like any Mac app
    3. It runs in your menu bar
  </Step>

  <Step title="Download a model">
    Open Terminal and run:

    ```bash theme={null}
    ollama pull llama3.1
    ```

    This downloads Meta's Llama 3.1 model (\~4.7GB). Other good options:

    * `ollama pull mistral` (faster, smaller)
    * `ollama pull codellama` (great for code)
    * `ollama pull mixtral` (more capable, larger)
  </Step>

  <Step title="Connect to Alter">
    1. Open Alter → Settings (`Cmd + ,`)
    2. Go to **API Keys** tab
    3. Under **Custom Provider**, select **Ollama**
    4. Make sure Ollama is running (check your menu bar)
    5. Toggle **Enable Custom Provider** ON
  </Step>

  <Step title="Start using it">
    1. Press `/` in Alter's prompt box
    2. Look for the **Custom** section
    3. Select your local model (e.g., "llama3.1")
    4. Ask anything – it runs entirely on your Mac!
  </Step>
</Steps>

### Option 2: LM Studio (More control)

LM Studio gives you a graphical interface to manage models.

<Steps>
  <Step title="Install LM Studio">
    Download from [lmstudio.ai](https://lmstudio.ai) and install it.
  </Step>

  <Step title="Download a model">
    1. Open LM Studio
    2. Browse the model catalog
    3. Download one that fits your needs and Mac's specs
    4. Start the local server (big "Start Server" button)
  </Step>

  <Step title="Connect to Alter">
    1. In Alter Settings → **API Keys**
    2. Select **LM Studio** as the provider
    3. LM Studio runs on `localhost:1234` by default
    4. Toggle **Enable Custom Provider** ON
  </Step>
</Steps>

## Real-world example

**Dr. Chen, physician:**

> "I need to analyze patient notes for research, but I can't use cloud AI due to HIPAA. With a local model, I can ask 'What patterns do you see in these symptoms?' and get AI assistance while keeping everything on my secure laptop."

**Alex, security researcher:**

> "I analyze malware reports and can't upload them anywhere. Running a local model means I can ask 'What indicators of compromise are mentioned?' without risking data exposure."

## Trade-offs to know about

<AccordionGroup>
  <Accordion icon="gauge" title="Speed vs Privacy">
    **Cloud models:** Faster, more capable
    **Local models:** Slower, but completely private

    A MacBook Pro with 16GB RAM can run small models smoothly. For larger models, you'll want 32GB+ RAM.
  </Accordion>

  <Accordion icon="brain" title="Capability differences">
    Local models are getting better every month, but cloud models (GPT-4, Claude) are still more capable for complex reasoning.

    **Best approach:** Use local models for sensitive data, cloud models for less sensitive complex tasks.
  </Accordion>

  <Accordion icon="hard-drive" title="Storage requirements">
    Models range from 4GB to 70GB+. Make sure you have enough disk space.

    **Good starter models:**

    * Llama 3.1 8B (\~4.7GB) – Fast, decent quality
    * Mistral 7B (\~4.1GB) – Good balance
    * Mixtral 8x7B (\~26GB) – Higher quality, needs more RAM
  </Accordion>
</AccordionGroup>

## Best practices for private workflows

<Tip>
  **Use specific Actions for sensitive tasks:** Create Alter Actions configured to always use your local model. That way you never accidentally use a cloud model for sensitive work.
</Tip>

<Tip>
  **Set up a privacy workspace:** Create a workspace just for sensitive documents. This keeps them organized and reminds you to use local models.
</Tip>

<Tip>
  **Verify it's working:** After setting up, disconnect from WiFi and try using Alter. If it still works with your local model, you know it's truly local.
</Tip>

## When to use what

| Scenario           | Recommendation                           |
| ------------------ | ---------------------------------------- |
| Medical records    | Local model (Ollama/LM Studio)           |
| Legal documents    | Local model                              |
| Security research  | Local model                              |
| Financial analysis | Local model or Alter Cloud with Pro plan |
| Creative writing   | Cloud models (better creativity)         |
| General questions  | Cloud models (faster)                    |
| Code assistance    | Either works well                        |

## Related resources

* [How to use your own API key](/how-to/use-byok) – Step-by-step setup guide
* [How to access settings](/how-to/settings-guide) – Open settings and configure providers
* [General FAQ](/references/general-faq) – Model and provider basics

***

**Ready to go private?** Start with Ollama and the Llama 3.1 model. It's free, easy to set up, and you'll have AI assistance that never leaves your Mac!