# What is MindMac?

## Introduction

[MindMac](https://mindmac.app/) is a feature-rich & privacy-first ChatGPT app for macOS to use OpenAI, Azure OpenAI,  GoogleAI with Gemini, Google Cloud Vertex AI with Gemini, Anthropic Claude, OpenRouter, MistralAI, Cohere, Perplexity, OctoAI and any local LLMs all in one place, designed for maximum productivity. Currently localization available in 15 languages but you can ask the bot in any languages.

For example with OpenAI, there are two different ways we can use all OpenAI models:

1. We can subscribe to ChatGPT Plus for $20 on a monthly basis and use it only on the web at <https://chat.openai.com/chat>.
2. Or, we can pay for the OpenAI API as you go and use app like MindMac on your Mac devices to access it. This option appears to be more cost-effective for a lot of users because they never end up spending more than $5 per month on the OpenAI API.

MindMac offers a variety of helpful features, and we are constantly working to add even more. If you want to stay updated on the features we're planning to add in the future, you can check out  our roadmap at [https://mindmac.canny.io](https://mindmac.canny.io/). Additionally, if there's a specific feature you'd like to see in MindMac, you can request it through Canny platform at <https://mindmac.canny.io/feature-requests>. We value your feedback and are dedicated to improving our product based on user suggestions.

## Why MindMac?

If you want:

* To use AI API to save cost
* A better UI with many more features to use AI API
* Ability to use multiple API providers at the same place
* To integrate seamlessly with macOS, such as the Inline mode, to leverage AI functionality in various applications
* An UI to use local LLMs
* An AI app that supports highly customizable keyboard shortcuts

Then, MindMac is a great option for your needs. MindMac is now being used daily by tens of thousands of users worldwide, and the user count continues to grow every day.

## Supported API providers

* [OpenAI](https://openai.com/) (Text, DALL·E 2, DALL·E 3, Vision)
* [Azure OpenAI](https://azure.microsoft.com/en-us/products/ai-services/openai-service)
* [Anthropic Claude](https://www.anthropic.com/) (Claude 2, Claude 3 Sonnet/Opus/Haiku, Claude Instant)
* [Google AI](https://deepmind.google/technologies/gemini) (with Gemini Pro and Gemini Pro Vision)
* [Google Cloud Vertex AI](https://cloud.google.com/vertex-ai) (with Gemini Pro and Gemini Pro Vision)
* [Mistral AI](https://mistral.ai/)
* [OctoAI](https://octo.ai/)
* [TogetherAI](https://www.together.ai/)
* [Groq](https://wow.groq.com/)
* [Anyscale](https://www.anyscale.com/endpoints)
* [Cohere](https://cohere.com/)
* [Perplexity](https://docs.perplexity.ai/)
* [OpenRouter](https://openrouter.ai/)
* Any other AI providers support OpenAI-compatible API
* Any local LLMs via [LMStudio](https://lmstudio.ai/)/[Ollama](https://ollama.ai/)/[GPT4All](https://gpt4all.io/index.html)

## Supported localization languages

* English
* Spanish
* French
* German
* European Portuguese
* Brazilian Portuguese
* Italian
* Arabic
* Japanese
* Korean
* Simplified Chinese&#x20;
* Traditional Chinese&#x20;
* Indonesian
* Thai
* Vietnamese

## Useful links

* Homepage: [https://mindmac.app](https://mindmac.app/)
* Twitter/X: <https://twitter.com/MindMacApp>
* Youtube demo videos: <https://www.youtube.com/@mindmac>
* Roadmap: [https://mindmac.canny.io](https://mindmac.canny.io/)
* Change log: <https://mindmac.canny.io/changelog>
* Feature request: <https://mindmac.canny.io/feature-requests>
* Documentation: [https://docs.mindmac.app](https://docs.mindmac.app/)
* Blog: [https://blog.mindmac.app](https://blog.mindmac.app/)


# Getting started

MindMac is freemium application, that means you can use it freely with some limitations and have option to upgrade to premium version.

### Add API key

After downloading and installing the app, you need to add your own API key to start exploring ChatGPT with MindMac. At the first time launch the app, you can follow below video to add quickly an API key.&#x20;

{% embed url="<https://www.youtube.com/watch?v=s6ysm3Z-dQ8>" %}

If you want to add more API providers, just going to [Add API key](/how-to.../add-api-key) page for more instructions.

To get API key from AI providers, check below articles.

* OpenAI: [Create OpenAI API Key](/how-to.../add-api-key/create-openai-api-key)
* Azure OpenAI: [Create Azure OpenAI API Key](/how-to.../add-api-key/create-azure-openai-api-key)
* Google AI: [Create Google AI API key](/how-to.../add-api-key/create-google-ai-api-key)
* Google Vertex AI: [Create Google Cloud Vertex AI API Key](/how-to.../add-api-key/create-google-cloud-vertex-ai-api-key)
* OpenRouter: [Create OpenRouter API Key](/how-to.../add-api-key/create-openrouter-api-key)
* Mistral AI: [Create Mistral AI API Key](/how-to.../add-api-key/create-mistral-ai-api-key)
* Perplexity: [Create Perplexity API Key](/how-to.../add-api-key/create-perplexity-api-key)
* Anyscale: [Create Anyscale API key](/how-to.../add-api-key/create-anyscale-api-key)

With the free version, you can try it freely with two limitations:

* Create up to 5 conversations
* Use Inline mode up to 20 times per day

### Activate license

In the case you want to upgrade to premium version, checkout at [here](https://mindmac.app/#pricing). Once payment has been successfully, Paddle (MindMac's payment gateway) will send you an email included a license key. Now, let go to [Activate license](/how-to.../activate-license) page to see how to activate the license on your Mac device.


# Features

## Main Window

<figure><img src="/files/wIaGaIRLRjIxAVorZVmM" alt=""><figcaption></figcaption></figure>

## Setting Window

TBD

## Inline Mode

TBD

##


# Activate license

Please check below video for more details.

{% embed url="<https://www.youtube.com/watch?v=B0q70XD1gtc>" %}


# Deactivate license

In the case you want to deactivate the license for some reasons, for example to use it on another machine, follow below steps to do so.

* Go to **MindMac** -> **Settings...** -> **Account** or press <mark style="color:purple;">**⌘**</mark> + <mark style="color:purple;">**,**</mark> to open Account Setting
* Click **Deactivate**

<figure><img src="/files/V4IMBUEsauXoKazTN78l" alt=""><figcaption></figcaption></figure>

* Click **Deactivate** to confirm

<figure><img src="/files/ss67SN5Pp9piQu9b0mbO" alt=""><figcaption></figcaption></figure>


# Add API key

MindMac supports multiple AI providers, not just OpenAI. In this article, you will learn how to add new API endpoint for a provider. You can also add multiple endpoints per provider as you want.

Go to **MindMac** -> **Settings...** -> **Account** or press <mark style="color:purple;">**⌘**</mark> + <mark style="color:purple;">**,**</mark> to open Account Setting.

<figure><img src="/files/E1a2mMoHRbxB9bcEg5Jq" alt=""><figcaption></figcaption></figure>

Select AI API provider, enter name, URL and API key, then click **Save** to finish.

<figure><img src="/files/RYlFtVigP7y4pCbuXEK7" alt=""><figcaption></figcaption></figure>

URL will be filled by default but for some providers like Azure OpenAI or Google Cloud Vertex AI, you might need to bring your own custom URL. For example, with Google Cloud Vertex AI, URL has format <https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/publishers/google/models>, where LOCATION and PROJECT\_ID is your own information. Check more on how to get API key and how to add Vertex AI key at [Create Google Cloud Vertex AI API Key](/how-to.../add-api-key/create-google-cloud-vertex-ai-api-key).

Check below articles to know how to get API key from AI providers.

* OpenAI: [Create OpenAI API Key](/how-to.../add-api-key/create-openai-api-key)
* Azure OpenAI: [Create Azure OpenAI API Key](/how-to.../add-api-key/create-azure-openai-api-key)
* Google AI: [Create Google AI API key](/how-to.../add-api-key/create-google-ai-api-key)
* Google Vertex AI: [Create Google Cloud Vertex AI API Key](/how-to.../add-api-key/create-google-cloud-vertex-ai-api-key)
* OpenRouter: [Create OpenRouter API Key](/how-to.../add-api-key/create-openrouter-api-key)
* Mistral AI: [Create Mistral AI API Key](/how-to.../add-api-key/create-mistral-ai-api-key)
* Perplexity: [Create Perplexity API Key](/how-to.../add-api-key/create-perplexity-api-key)


# Create OpenAI API Key

If you don't have an account, let's create one at <https://platform.openai.com/signup>.&#x20;

### Step 1: Add payment details

If you don't have free credit, you might need to add payment detail and add credit balance to start using OpenAI API. Go to <https://platform.openai.com/account/billing/overview> and click "Add payment details".&#x20;

Otherwise, if you already have credit balance, just skip this step and go to [#step-6-go-to-https-platform.openai.com-api-keys-and-click-create-new-secret-key](#step-6-go-to-https-platform.openai.com-api-keys-and-click-create-new-secret-key "mention").

<figure><img src="/files/xnecIhy2LyDlwvwcTAZf" alt=""><figcaption></figcaption></figure>

### Step 2: Select payment type

Choose your best fit payment type. Let try with "Individual" if you plan to use the API personally.

<figure><img src="/files/coQ4AMuakCZzefnY5MkV" alt=""><figcaption></figcaption></figure>

### Step 3: Enter card information

<figure><img src="/files/4QVQ3J5xuYKuu8DmQqf4" alt=""><figcaption></figcaption></figure>

### Step 4: Add credit balance

<figure><img src="/files/xNTsK2GEFDMpm4Cmkanl" alt=""><figcaption></figcaption></figure>

### Step 5: Enter amount

<figure><img src="/files/en5XRGoY5RlD6SAqv4DH" alt=""><figcaption></figcaption></figure>

### Step 6: Go to <https://platform.openai.com/api-keys> and click "Create new secret key"

<figure><img src="/files/ilCOH5m6INDZdxyEj1jC" alt=""><figcaption></figcaption></figure>

### Step 7: Enter key name and click "Create secret key"

<figure><img src="/files/86dRFFmT5unpge27eajU" alt=""><figcaption></figcaption></figure>

### Step 8: Copy the API key. Keep it secret.

Please notice that you won't be able to view the key again through your OpenAI account for security reasons. If you lose this secret key, you'll need to generate a new one.

<figure><img src="/files/ST4fzFSWTWLYuVEu76pV" alt=""><figcaption></figcaption></figure>

Now go to [Add API key](/how-to.../add-api-key) and use copied API key to create new API endpoint.


# Create Azure OpenAI API Key

### Step 1: Create Azure OpenAI resource

Go to <https://portal.azure.com/#home>, and click **Create a resource**.

<figure><img src="/files/1G2fkW3O3KlwNb3BJpLB" alt=""><figcaption></figcaption></figure>

Search for **Azure OpenAI** and select it.

<figure><img src="/files/4f4nBTCEFkH44jXm89h6" alt=""><figcaption></figcaption></figure>

Select plan and click **Create**.

<figure><img src="/files/RMwqsZKdmaZO3LeC92xi" alt=""><figcaption></figcaption></figure>

Enter below information and click **Next**.

<figure><img src="/files/u6ZZqxH8qYuz3B7uu3xd" alt=""><figcaption></figcaption></figure>

Configure your network preference and click **Next**.

<figure><img src="/files/11fd305XP8vzwzfn6Gio" alt=""><figcaption></figcaption></figure>

Click **Next**.

<figure><img src="/files/a3k81osJ9XxHwcOBGqCy" alt=""><figcaption></figcaption></figure>

Review and click **Create**.

<figure><img src="/files/hEfo5kMZ1HoqD9VWvMrd" alt=""><figcaption></figcaption></figure>

It may take some minutes to deploy new resource.&#x20;

### Step 2: Grab API key, location & endpoint URL

When the resource has been deployed successfully, go back to <https://portal.azure.com/#home> and click on that resource.

<figure><img src="/files/OFJemRCSXSmDDI5aJZl1" alt=""><figcaption></figcaption></figure>

On tab **Overview**, copy key, location/region and endpoint URL. You will need them later.

<figure><img src="/files/ToUDmC8CLjTVvMkRmNk7" alt=""><figcaption></figcaption></figure>

### Step 3: Create new deployment

Switch to tab **Model deployments** and click **Manage Deployments**. You will be redirect to Azure OpenAI Studio.

<figure><img src="/files/ANbd94bQent3XbMIomD1" alt=""><figcaption></figcaption></figure>

Click **Create new deployments**.

<figure><img src="/files/LDuJpdCzbuV6jpzOZUx1" alt=""><figcaption></figcaption></figure>

Enter deployment model information and click **Create**. It may take some minutes to start a deployment.

<figure><img src="/files/DZJ4G2oOogdnhccfxCCV" alt=""><figcaption></figcaption></figure>

Copy the deployment name.

<figure><img src="/files/UoJ4zbVVnNivVv9FsQCp" alt=""><figcaption></figcaption></figure>

### Step 4: Construct full endpoint URL

Full endpoint URL has this format:&#x20;

> {your-endpoint}/openai/deployments/{deployment-id}/chat/completions?api-version={api-version}

where:

* your-endpoint: is the endpoint copied from [#step-2-grab-api-key-location-and-endpoint-url](#step-2-grab-api-key-location-and-endpoint-url "mention")&#x20;
* deployment-id: deployment name from [#step-3-create-new-deployment](#step-3-create-new-deployment "mention")
* api-version: use 2023-05-15 for now

For example: <https://mindmac.openai.azure.com/openai/deployments/mindmac-gpt-35-turbo/chat/completions?api-version=2023-05-15>.

Now use this full URL and the key from [#step-2-grab-api-key-location-and-endpoint-url](#step-2-grab-api-key-location-and-endpoint-url "mention") to create new API endpoint in MindMac. More details at [Add API key](/how-to.../add-api-key).


# Create Google AI API key

### Step 1: Create API key

Go to <https://makersuite.google.com/app/apikey> and click **Create API key in existing project**

<figure><img src="/files/PqGJrAZkXfIRZvnKqza5" alt=""><figcaption></figcaption></figure>

### Step 2: Select a project and click Create new API key

<figure><img src="/files/f9XIHrsh6q0z4EMSLcjH" alt=""><figcaption></figcaption></figure>

### Step 3: Copy the API key. Keep it secret.

<figure><img src="/files/8D6pxmmLQvJXzEFItFaF" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/bR8VqjZvmMenH3piKdlp" alt=""><figcaption></figcaption></figure>

Now go to [Add API key](/how-to.../add-api-key) and use copied API key to create new endpoint.


# Create Google Cloud Vertex AI API Key

### Step 1: Create new project

Go to <https://console.cloud.google.com/projectcreate>, enter project name and click **Create.**

<figure><img src="/files/mxEXEn7AxOXITIIHRdZ9" alt=""><figcaption></figcaption></figure>

### Step 2: Create new service account

Go to [https://console.cloud.google.com/iam-admin/serviceaccounts](https://console.cloud.google.com/iam-admin/serviceaccounts?authuser=1), select the project just created in [#step-1-create-new-project](#step-1-create-new-project "mention").

Click **Create Service Account**.

<figure><img src="/files/2klFbRx7q16cHXSy2LnB" alt=""><figcaption></figcaption></figure>

Enter name, description for service account and click **Create and Continue**.

<figure><img src="/files/cXlThvQACLuJRToi3zsa" alt=""><figcaption></figcaption></figure>

Click **Select a role** and search for **Vertex AI Service Agent**, then select it.

<figure><img src="/files/3nZblXRZ5bdO6BOReKBP" alt=""><figcaption></figcaption></figure>

Click **Continue**.

<figure><img src="/files/LCO2dGwDe2hGmrxSkkk2" alt=""><figcaption></figcaption></figure>

Click **Done**.

<figure><img src="/files/qVEspVhwnesbo9TUXrqF" alt=""><figcaption></figcaption></figure>

### Step 3: Generate new key

Click on the service account just created, then select **Keys** tab.

<figure><img src="/files/2o2mwmBhDpWWmq88Pmql" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/4sF98vwQGzUlzT8YmIc4" alt=""><figcaption></figcaption></figure>

Click **Add Key** and select **Create new key**.

<figure><img src="/files/UYLvq3bJKn2tniOlyWAb" alt=""><figcaption></figcaption></figure>

Select **JSON** and click **Create**.

<figure><img src="/files/a7GBmD0T0p8fQV4I2e0Q" alt=""><figcaption></figcaption></figure>

A JSON file will be downloaded into your machine. Keep that file secret. Open that file and copy whole file content.

<figure><img src="/files/R8LLxSaiafoaSejuHTtY" alt=""><figcaption></figcaption></figure>

Key file content looks like below:

```
{
  "type": "service_account",
  "project_id": "mindmac-300128",
  "private_key_id": "12efb3bcc7u7dbf111842f27415d37454c6d6fa012",
  "private_key": "-----BEGIN PRIVATE KEY-----\nMIIE...8XI3A=\n-----END PRIVATE KEY-----\n",
  "client_email": "mindmac-service-account@mindmac-300128.iam.gserviceaccount.com",
  "client_id": "208453549200948148846",
  "auth_uri": "https://accounts.google.com/o/oauth2/auth",
  "token_uri": "https://oauth2.googleapis.com/token",
  "auth_provider_x509_cert_url": "https://www.googleapis.com/oauth2/v1/certs",
  "client_x509_cert_url": "https://www.googleapis.com/robot/v1/metadata/x509/mindmac-service-account%40mindmac-300128.iam.gserviceaccount.com",
  "universe_domain": "googleapis.com"
}
```

### Step 4: Enable Vertex AI API

Go to <https://console.cloud.google.com/marketplace/product/google/aiplatform.googleapis.com>, select the project just created and click **Enable** to activate Vertex AI API.

<figure><img src="/files/gQnhrmC4gXusOQ8BqPWd" alt=""><figcaption></figcaption></figure>

### Step 5: Grab information to construct API URL

Go to <https://console.cloud.google.com/vertex-ai>, select project just created and take note region name.

<figure><img src="/files/xDsdGomNU88WDEyWI07b" alt=""><figcaption></figcaption></figure>

Go to <https://console.cloud.google.com/welcome>, select project just created and copy the project ID.

<figure><img src="/files/icBukE62h4xdv6dnVyqp" alt=""><figcaption></figcaption></figure>

Vertex AI API URL has format:&#x20;

```
https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/LOCATION/publishers/google/models
```

* LOCATION is the region name above, for example **us-central1**
* PROJECT\_ID is the ID of the project above

In this case it should be:

```
https://us-central1-aiplatform.googleapis.com/v1/projects/mindmac-300128/locations/us-central1/publishers/google/models
```

***

### Add Google Vertex AI key

Go to **MindMac** -> **Settings...** -> **Account** or press <mark style="color:purple;">**⌘ + ,**</mark> to open Account Setting.

<figure><img src="/files/pOxx0grWm6zgadWDp24W" alt=""><figcaption></figcaption></figure>

Select **Vertex AI** provider, enter name and your custom URL from [#step-5-grab-information-to-construct-api-url](#step-5-grab-information-to-construct-api-url "mention"), paste Service Account key content from [#step-2-create-new-service-account](#step-2-create-new-service-account "mention") and click **Save** to finish.


# Create OpenRouter API Key

Create an account if you don't have one at [https://openrouter.ai](https://openrouter.ai/).

### Step 1: Add credit balance

If you don't have any credit balance, go to <https://openrouter.ai/account#credits> and click "**Add Credits**", then follow instructions to enter your billing information.

Otherwise, skip this step and go to [#step-2-create-a-api-key](#step-2-create-a-api-key "mention").

<figure><img src="/files/IC4bJ0iVauv3Mg96fGpO" alt=""><figcaption></figcaption></figure>

### Step 2: Create a API key

Visit <https://openrouter.ai/keys> and click "**Create Key**".

<figure><img src="/files/shFxkdmHDiWjZfYYiTPU" alt=""><figcaption></figcaption></figure>

### Step 3: Enter key name

In this step, you can set credit limit for this specific key to avoid overuse or keep it blank for unlimited use.

<figure><img src="/files/3H5kS495282jRGYtspVz" alt=""><figcaption></figcaption></figure>

### Step 4: Copy the API key. Keep it secret.

Please notice that you won't be able to view the key again for security reasons. If you lose this secret key, you'll need to generate a new one.

<figure><img src="/files/HDMTTEd4cTLqrLiAibOS" alt=""><figcaption></figcaption></figure>

Now go to [Add API key](/how-to.../add-api-key) and use copied API key to create new endpoint.


# Create Mistral AI API Key

Create an account if you don't have one at <https://auth.mistral.ai/ui/registration>. However, you might be in the waitlist to use the platform. So you can contact [support@mistral.ai](<mailto:support@mistral.ai >) to ask for early access but not sure it always work, unfortunately.

<figure><img src="/files/7cNP4zhWRRH1bvZ7aPVR" alt=""><figcaption></figcaption></figure>

Once you have an active account, follow below steps to get the API key.

### Step 1: Add billing information

Go to <https://console.mistral.ai/billing/> and follow instructions to add billing information and add credit balance.

### Step 2: Create a API key

Go to <https://console.mistral.ai/user/api-keys/> and click "**Generate a new key**".

<figure><img src="/files/dliGtb0kgL4LuqcNCzD3" alt=""><figcaption></figcaption></figure>

### Step 3: Copy the API key. Keep it secret.

Please notice that you won't be able to view the key again for security reasons. If you lose this secret key, you'll need to generate a new one.

<figure><img src="/files/N3GcjzD6NkOoUmWsDU8U" alt=""><figcaption></figcaption></figure>

Now go to [Add API key](/how-to.../add-api-key) and use copied API key to create new endpoint.


# Create Perplexity API Key

If you don't have an account, create a new one at [https://www.perplexity.ai](https://www.perplexity.ai/).

### Step 1: Add billing information

Go to <https://www.perplexity.ai/settings/api>

<figure><img src="/files/I6NDvAc8rZptwy78G2rN" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/Jm8ofzvJzrZTygJGRZKK" alt=""><figcaption></figcaption></figure>

### Step 2: Buy credits

In the section **Available Credits**, click **Buy Credits** and enter the amount.

<figure><img src="/files/gupcZfYuu89WXWKAaOb7" alt=""><figcaption></figcaption></figure>

### Step 3: Generate API Key

In the **API Keys** sections, click **Generate**

<figure><img src="/files/RWUjjdFAmepM4qBg9DZV" alt=""><figcaption></figcaption></figure>

### Step 4: Copy the API key. Keep it secret.

<figure><img src="/files/PeYRAUhcReXLxFctjBHw" alt=""><figcaption></figcaption></figure>

Now go to [Add API key](/how-to.../add-api-key) and use copied API key to create new endpoint.


# Create Replicate API Key

If you don't have an account, create a new one at <https://replicate.com/signin>.

### Step 1: Add billing information

Go to <https://replicate.com/account/billing> and follow instruction to add billing information.

### Step 2: Create API key

* Go to <https://replicate.com/account/api-tokens>
* Enter a key name
* Click **Create token**

<figure><img src="/files/RKSMBXmsMbO7bRxeinvO" alt=""><figcaption></figcaption></figure>

* Click copy button to get the API key. Keep it secret.

<figure><img src="/files/xChKUU8VkcmOQpx89xoc" alt=""><figcaption></figcaption></figure>


# Create Anyscale API key

If you don't have an account, create a new one at <https://app.endpoints.anyscale.com/welcome>.

Once you have an account, go to <https://app.endpoints.anyscale.com/credentials> and follow 2 steps below to generate new API key.

<figure><img src="/files/PHw8CELSZGm7dlT3Df91" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/iRxvDdYKNREPiqEctQTd" alt=""><figcaption></figcaption></figure>

Now go to [Add API key](/how-to.../add-api-key) and use copied API key to create new endpoint.


# Add Ollama endpoint

## Prerequisite

* Ollama version: 0.1.19 or later
* Running at least one model under Ollama

## Add endpoint

Go to **MindMac** -> **Settings...** -> **Account** or press <mark style="color:purple;">**⌘**</mark> + <mark style="color:purple;">**,**</mark> to open Account Setting.

<figure><img src="/files/E1a2mMoHRbxB9bcEg5Jq" alt=""><figcaption></figcaption></figure>

* Select **Ollama** as provider
* Enter a name
* If you don't change any Ollama configuration, just use <http://localhost:11434/api/chat> as URL
* Enter a random text in API Key text area
* Click **Save** to finish

<figure><img src="/files/Nd0y3smSropHgGxWkfbE" alt=""><figcaption></figcaption></figure>

***

## Customize Context Length (Optional)

Currently Ollama does not provide exact context length for each model, so if you want to control **max tokens** (or context length) parameter,  you might need to enter that value manually for every Ollama model. Follow below steps to do so.

Go to **MindMac** -> **Settings...** -> **Account** or press <mark style="color:purple;">**⌘**</mark> + <mark style="color:purple;">**,**</mark> to open Account Setting.

Click on the brain icon to show Ollama model list.

<figure><img src="/files/iWk0Y0FE6i803FD66aAQ" alt=""><figcaption></figcaption></figure>

Click on the edit button.

<figure><img src="/files/s0AzEdXF9JLc39MpArXc" alt=""><figcaption></figcaption></figure>

Enter max tokens value for that model and click **Save** to finish.

<figure><img src="/files/zzAGb8WziGgA1c7PmVqm" alt=""><figcaption></figcaption></figure>

To get max tokens value for each model, you can find in configuration file (**config.json** for example) of that model on Hugging Face. For instance, with the model [Mixtral-8x7B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1) , you can go to <https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1/blob/main/config.json> and grab the value **max\_position\_embeddings** as its context length. Kindly be aware that there is no standard specifying which key determines the context length of a model in the model configuration file.

<figure><img src="/files/h8sezoxAHS1iLnckM4zY" alt=""><figcaption></figcaption></figure>


# Internet Browsing

Almost AI models now have training cutoff date, so this feature allows you to access the latest data from the internet to resolve that issue.

## Prerequisite

To use Internet browsing feature, you need to configure a search provider. Currently MindMac supports 5 providers:

| Name                                                                                    | Free                   | Paid                                                                    | Pricing page                           | Get API Key                                                                   |
| --------------------------------------------------------------------------------------- | ---------------------- | ----------------------------------------------------------------------- | -------------------------------------- | ----------------------------------------------------------------------------- |
| [Google Programmable Search Engine](https://programmablesearchengine.google.com/about/) | 100 queries per day    | Additional requests cost $5 per 1000 queries, up to 10k queries per day |                                        | [Get Google Search Key](/how-to.../internet-browsing/get-google-search-key)   |
| [Tavily](https://tavily.com/)                                                           | 1000 queries per month | **Startup** plan costs 100$/month for 15000 API calls                   | [Pricing](https://tavily.com/#pricing) | [Get Tavily Key](/how-to.../internet-browsing/get-tavily-key)                 |
| [You Search](https://api.you.com/)                                                      | 1000 queries per month | **Explorer AI** plan costs 100$/month for 14200 API calls               | [Pricing](https://api.you.com/plans)   | [Get You.com Search Key](/how-to.../internet-browsing/get-you.com-search-key) |
| [Serper](https://serper.dev/)                                                           | First 2500 queries     | After free calls, 50$ for 50000 queries                                 | [Pricing](https://serper.dev/)         | [Get Serper Key](/how-to.../internet-browsing/get-serper-key)                 |
| [SerpApi](https://serpapi.com/)                                                         | 100 queries per month  | **Developer** plan costs 50$/month for 5000 searches                    | [Pricing](https://serpapi.com/pricing) | [Get SerpApi Key](/how-to.../internet-browsing/get-serpapi-key)               |

## Configuration

Enable **Internet Browsing** and click **Configure** to add new search provider.

<figure><img src="/files/XIIrUwGD0WTZn4aud4JP" alt=""><figcaption></figcaption></figure>

Select search provider, enter name, search engine ID (for Google Programmable Search) and API key. Click Save to finish.&#x20;

<figure><img src="/files/NlWk9xsYSILhsfMhMYSk" alt=""><figcaption></figcaption></figure>

## Try it

Ask something to get the latest news. You can click **References** to view list of information sources.

<figure><img src="/files/AkQiFYBOqbXjtv9U6251" alt=""><figcaption></figcaption></figure>

## Manage search providers

Go to **MindMac** -> **Settings...** -> **Internet Browsing** or press <mark style="color:purple;">**⌘**</mark> + <mark style="color:purple;">**,**</mark> to open Internet Browsing Setting.

You will see list of search providers. Here you can edit a search provider (rename, update API key), add a new provider or delete a provider.

<figure><img src="/files/NvpsszyhZvJV4OcRaK6l" alt=""><figcaption></figcaption></figure>

To add a new search provider:

* Click on the plus button on the bottom
* Select provider, enter name, API key, search engine ID (for Google Programmable Search)
* Click **Save** to finish

After that, you can switch between search providers on the main UI window or using shortcut <mark style="color:purple;">⇧⌘L</mark>.

<figure><img src="/files/ScZ9znoGhEvhAHMb2Xyb" alt=""><figcaption></figcaption></figure>

## Useful keyboard shortcuts

* Enable Internet Browsing feature: <mark style="color:purple;">⇧⌘B</mark>
* Switch search provider: <mark style="color:purple;">⇧⌘L</mark>

These keyboard shortcuts can be customized in **MindMac** -> **Settings...** -> **Shortcut**.


# Get Google Search Key

To use Google Programmable Search Engine, you need Search Engine ID and Custom Search API key. Follow these below steps to do so.

### Step 1: Go to [https://programmablesearchengine.google.com/about](https://programmablesearchengine.google.com/about/) and click "Get Started"

<figure><img src="/files/2ymEKkUOIeDPjePgmKpX" alt=""><figcaption></figcaption></figure>

### Step 2: Create new search engine

<figure><img src="/files/F5uddYRsyRdwIDas9vLE" alt=""><figcaption></figcaption></figure>

### Step 3: Click "Customize"

<figure><img src="/files/FVzAPepkftQXd2coZSNX" alt=""><figcaption></figcaption></figure>

### Step 4: Copy Search Engine ID

<figure><img src="/files/HD9YK7B0E0iI6GPuy1yX" alt=""><figcaption></figcaption></figure>

### Step 5: Keep scrolling to "Programmatic Access" section and click "Get Started" of "Custom Search JSON API"

<figure><img src="/files/CB0S1c5NiOM9DYAwFOtV" alt=""><figcaption></figcaption></figure>

### Step 6: Click "Get a Key"

<figure><img src="/files/rY7ltxs1rO4RLRxqL5z3" alt=""><figcaption></figcaption></figure>

### Step 7: Select a project and click "Next"

<figure><img src="/files/KDCcPiaToCrkdYjJchdk" alt=""><figcaption></figcaption></figure>

### Step 8: Click "Show Key"

<figure><img src="/files/cjZCJwJn01BPdXi0jjyK" alt=""><figcaption></figcaption></figure>

### Step 9: Copy API key and keep it secret

<figure><img src="/files/8JwJUzMrUgpav4HictMB" alt=""><figcaption></figcaption></figure>

Now you have Search Engine ID in step 4 and API key in step 9 to configure search provider for MindMac.


# Get Tavily Key

### Step 1: Go to <https://app.tavily.com/sign-in> and click "Sign In"

<figure><img src="/files/q63q80E0bWqXN9mOzz4d" alt=""><figcaption></figcaption></figure>

### Step 2: Select Sign In method. Let try with Google

<figure><img src="/files/VmQ07Dyun0WWLfnG5YKt" alt=""><figcaption></figcaption></figure>

### Step 3: Select one of your Google accounts

<figure><img src="/files/URofbUbGcETIx1q4KSI0" alt=""><figcaption></figcaption></figure>

### Step 4: You will be redirect to <https://app.tavily.com/home>. Go to "API Key" section and copy the key. Keep it secret.

<figure><img src="/files/hD61RI5GiaVGPdPuIdOS" alt=""><figcaption></figcaption></figure>


# Get You.com Search Key

### Step 1: Go to <https://api.you.com/api-key>, create an account if you don't have one

<figure><img src="/files/kXbDnlY8QPg2SG6WgexG" alt=""><figcaption></figcaption></figure>

### Step 2: If you already have an account, sign in at <https://api.you.com/api-key>

<figure><img src="/files/ofP2kIZoqxwyB1X1IIG3" alt=""><figcaption></figcaption></figure>

### Step 3:  Click "Generate" of Web Search

<figure><img src="/files/AlQqWL8B28wFIfqhXEyJ" alt=""><figcaption></figcaption></figure>

### Step 4: Copy search API key. Keep it secret.

<figure><img src="/files/GXnqjzisDtwq5U6JTXvI" alt=""><figcaption></figcaption></figure>


# Get Serper Key

### Step 1: Create an account at <https://serper.dev/signup> if you don't have one

<figure><img src="/files/q5YHg4eNukyScQb2BECD" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/91DRWjS6dNlTX7x5TX16" alt=""><figcaption></figcaption></figure>

### Step 2: Sign In at <https://serper.dev/login>

<figure><img src="/files/wOlQX2ZbIQlhqsaOu0sH" alt=""><figcaption></figcaption></figure>

### Step 3: In the dashboard <https://serper.dev/dashboard>, select "API Key" section

<figure><img src="/files/ehAQyNbMCmfX3KGHctiZ" alt=""><figcaption></figcaption></figure>

### Step 4: Copy the API key. Keep it secret.

<figure><img src="/files/9KKe2hgIDfwKrw309cX9" alt=""><figcaption></figcaption></figure>


# Get SerpApi Key

### Step 1: Create an account at <https://serpapi.com/users/sign_up> if you don't have one.

<figure><img src="/files/MoKJp3UCEcHpE3hED9Mk" alt=""><figcaption></figcaption></figure>

### Step 2: Sign In and go to <https://serpapi.com/dashboard>, then copy the key at "Your Private API Key" section. Keep it secret.

<figure><img src="/files/LbVYWYDyEs0dhfs22Rs2" alt=""><figcaption></figcaption></figure>


# Use local LLM in MindMac

You can use local LLMs with MindMac via [LMStudio](https://lmstudio.ai/), [GPT4All](https://gpt4all.io/index.html) or [Ollama](https://ollama.ai/)+[~~LiteLLM~~](https://github.com/BerriAI/litellm). Please check below videos for more details.

* [How to use local LLMs in MindMac by using LMStudio?](https://www.youtube.com/watch?v=3KcVp5QQ1Ak)
* [How to use local LLMs with GPT4All & MindMac?](https://www.youtube.com/watch?v=4a6cIzDjh30)
* [H~~ow to use Ollama with MindMac using LiteLLM?~~](https://www.youtube.com/watch?v=bZfV70YMuH0\&t=1s)

### LMStudio

{% embed url="<https://www.youtube.com/watch?v=3KcVp5QQ1Ak>" %}

### GPT4All

{% embed url="<https://www.youtube.com/watch?v=4a6cIzDjh30>" %}

### Ollama+~~LiteLLM~~

{% embed url="<https://www.youtube.com/watch?t=1s&v=bZfV70YMuH0>" %}

Now you can use Ollama without LiteLLM. Please visit [Add Ollama endpoint](/how-to.../add-ollama-endpoint).


# Change App Icon

Go to **MindMac** -> **Settings...** -> **Configuration** or press <mark style="color:purple;">**⌘**</mark> + <mark style="color:purple;">**,**</mark> to open Configuration Setting.

In **Appearance** section, select the icon you want. It might be necessary to restart the app for changes to take effect in Spotlight search and the app switcher.

<figure><img src="/files/9IQYmshU3NLm10salNYL" alt=""><figcaption></figcaption></figure>


# Manage Models

## Add new model

In some cases, you might want to add your own model.  Just follow below steps to do so.

<div data-full-width="false"><figure><img src="/files/lrCIxIaPbXACvBAdIbqu" alt=""><figcaption></figcaption></figure> <figure><img src="/files/Je3bFVWfYD2WUdQnzn4w" alt=""><figcaption></figcaption></figure></div>

<div data-full-width="false"><figure><img src="/files/qW6VVyNpDTrDbvLejaUN" alt=""><figcaption></figcaption></figure> <figure><img src="/files/yAD4GGN3I9sMkuADq5mh" alt=""><figcaption></figcaption></figure></div>

## Update model

MindMac provides ability to modify model configuration: model name, model code, url, prompt price, completion or max tokens (context length). Please follow steps in below images to do so.

<div data-full-width="false"><figure><img src="/files/P5KxMklpXIrIzlKY84pu" alt=""><figcaption></figcaption></figure> <figure><img src="/files/Bq6vLancKcr80EKO8vlX" alt=""><figcaption></figcaption></figure> <figure><img src="/files/ZB8rloUxmAcpWR82flpc" alt=""><figcaption></figcaption></figure></div>

## Show or hide model from model list

You can control model list on main chat by following two below steps.

<div align="center" data-full-width="false"><figure><img src="/files/6RDTmFTMRqizgIpZ6b7J" alt=""><figcaption></figcaption></figure> <figure><img src="/files/mc8KNxmUu4mupYsBKAuk" alt=""><figcaption></figcaption></figure></div>


# License code does not match product

When you try to activate the license code but get this error **License code does not match product**, that means that code has been activated on other devices.&#x20;

At present, the Basic plan enables activation on a single device, the Personal plan allows activation on two devices, and the Standard plan supports activation on three devices. Therefore, if you wish to activate a license on additional devices, you will need to either deactivate the license on one of your existing devices or purchase another license. Please refer to the following steps to carry out these actions.

* Go to **MindMac** -> **Settings...** -> **Account** or press <mark style="color:purple;">**⌘**</mark> + <mark style="color:purple;">**,**</mark> to open Account Setting
* Click **Deactivate**

<figure><img src="/files/V4IMBUEsauXoKazTN78l" alt=""><figcaption></figcaption></figure>

* Click **Deactivate** to confirm

<figure><img src="/files/ss67SN5Pp9piQu9b0mbO" alt=""><figcaption></figcaption></figure>

From now, you can activate the license again on another device.


# Cannot open the app

In the case you cannot launch the app, it might be crashed somehow on your end. You may need to clean your MindMac data and reinstall the app. Please follow below steps to do that.

* Open Terminal app
* Remove the folder <mark style="color:yellow;">\~/Library/Application\ Support/app.mindmac.macos</mark> by command line:

```
rm -rf ~/Library/Application\ Support/app.mindmac.macos/
```

* Remove the folder <mark style="color:yellow;">\~/Library/Application\ Support/MindMac/</mark> by command line:

```
rm -rf ~/Library/Application\ Support/MindMac/
```

* Drag MindMac from Applications folder to Trash
* Reinstall MindMac


# Invalid URL or Key issue

At the very first time you add new API endpoint, you may encounter issue **Invalid URL or key:**

<figure><img src="/files/eySCU0VWnkeZCJY6LRQo" alt=""><figcaption><p>Invalid URL or key issue when add new API endpoint</p></figcaption></figure>

The first and most popular reason is that you have not included billing payment for the ChatGPT API. There is a difference between the ChatGPT API, which is required to be used in MindMac, and ChatGPT Plus. You can find further information [here](https://help.openai.com/en/articles/6950777-what-is-chatgpt-plus).&#x20;

> **Is the ChatGPT API included in the ChatGPT Plus subscription?**<br>
>
> *No, the ChatGPT API and ChatGPT Plus subscription are billed separately. The API has its own pricing, which can be found at* [*https://openai.com/pricing*](https://openai.com/pricing)*. The ChatGPT Plus subscription covers usage on chat.openai.com only and costs $20/month.*

To address this issue, simply **go to your** [**OpenAI billing dashboard**](https://platform.openai.com/account/billing/overview) and **add your billing information.**

Other reasons may occur when you input an incorrect website address or an incorrect API key (for example, if you accidentally copied it wrong). Additionally, your free credit may have expired. To resolve these issues, please take a moment to carefully double-check all of this data.


# Common questions

### License Key vs API Key

To activate the premium features of MindMac, a License Key is required while for interacting with ChatGPT through OpenAI's API, an API Key is needed. In order to utilize the app, an API Key is mandatory while a License Key is solely required for accessing the premium features.

### How does license work?

The license is a perpetual license. One license key allows you to use the current app version forever. You are also eligible for one year of free updates.&#x20;

After one year, if you want to get more update versions, you can renew the license for another year with 40% discount of the price you bought it.\
\
If you don't want to renew your license after one year, you can continue to use the last version of MindMac that you have for as long as you want.

### Do I need to pay for ChatGPT Plus to use MindMac?

Actually no. MindMac use ChatGPT API which is separate from ChatGPT Plus, so you don't need to subscribe ChatGPT Plus. More about the differences between them can be found [here](https://help.openai.com/en/articles/6950777-what-is-chatgpt-plus).

### Can I get a refund if I don't like it?

Certainly. We offer a 14-day money-back guarantee. If you're not happy, let us know within 14 days and we'll issue a full refund with no question asked.

### Does MindMac support Windows/Linux?

Currently, MindMac is only compatible with macOS. However, we continuously assess the requirements of our users, and we may explore the possibility of expanding to other platforms such as Windows or Linux in the future. Keep an eye out for any upcoming updates.


