# FAQ on GPT for Docs ## How do I install the GPT for Docs add-on? Install from the Google Workspace Marketplace: 1. Go to the [installation page](https://workspace.google.com/marketplace/app/gpt_for_sheets_%E2%80%93_autosheet/677318054654). 2. Click **Install**. 3. Follow the steps until the end. [Learn more.](https://autosheet.com/docs/gpt-for-sheets/setup/install) ## Which browsers are compatible with GPT for Docs? GPT for Docs is compatible with the following browsers: - Google Chrome (recommended for the best experience) - Microsoft Edge (only for Windows users) - Mozilla Firefox - Safari (only for Mac users) ## What permissions are required by the GPT for Docs add-on? See [Security and privacy FAQ](https://autosheet.com/docs/resources/security-and-privacy-faq#google-permissions-for-gpt-for-sheets-and-docs). ## I have installed GPT for Docs, how do I get started? Now that GPT for Docs is installed, you can launch it from the Google Docs **Extensions** menu and start editing or generating content with the add-on. ## What is the pricing of GPT for Docs? Autosheet uses [subscription pricing](https://autosheet.com/docs/resources/pricing/subscription). You pay a monthly fee based on the number of seats. Each seat includes $25 of usage per month. If you need more usage, you can buy extra-usage credits that are shared across your team and expire 12 months after your last top-up. Some existing spaces remain on legacy [pay-as-you-go pricing](https://autosheet.com/docs/resources/pricing/pay-as-you-go), where you buy credits and consume them as you go. To negotiate a custom plan, [contact sales](https://support.autosheet.com/hc/en-us/requests/new). ## How is my data processed? Is it used to train models? - We do not read or store any data from your document that you do not use as an input. - If you use a **connected** model, we do not log or store any of the inputs submitted or outputs received. - Whether you use a built-in or connected model, Talarian does not use your inputs and outputs to train language models. For more information, see the [Security and privacy FAQ](https://autosheet.com/docs/resources/security-and-privacy-faq). ## Do I need my own API key? Usually, no. Many models are [available as built-in models](https://autosheet.com/docs/resources/models/ai-providers-and-models#tk) with both [subscription](https://autosheet.com/docs/resources/pricing/subscription) and [legacy pay-as-you-go](https://autosheet.com/docs/resources/pricing/pay-as-you-go) pricing. Some custom Enterprise plans require an API key or endpoint. ## Can I use my own API key? With legacy pay-as-you-go pricing, yes. With subscription pricing, yes if you're on the Business or Enterprise plan. API keys are not available on the Standard plan. Using API keys offers several benefits: - **More models.** Access a [wider range of models from a bigger pool of AI providers](https://autosheet.com/docs/resources/models/ai-providers-and-models#byok) than what is available [without an API key](https://autosheet.com/docs/resources/models/ai-providers-and-models#tk). - **More control.** Monitor and manage AI usage and costs directly on the AI provider's platform. - **Improved privacy.** We do not log inputs and outputs when you use a connected model. For more information, see our [security and privacy FAQ](https://autosheet.com/docs/resources/security-and-privacy-faq#how-is-my-data-processed). - **Reduced costs.** If you're on a [pay-as-you-go space](https://autosheet.com/docs/resources/pricing/pay-as-you-go), using an API key reduces costs. You pay the model usage fees directly to the AI provider, while paying a reduced service fee for Autosheet. To estimate the total cost, compare the [Autosheet charges](https://autosheet.com/docs/resources/pricing/how-credits-are-consumed) with the AI provider's fees. Connecting an API key does not guarantee a lower total bill on every plan. ## Why is my response cut or incomplete? Models have a maximum token limit, which is the maximum total number of tokens that can be used in both the input and the response. This means models may not provide complete responses to long queries or when a large portion of text is selected as input. To stay within the token limit, you can: - Highlight a smaller portion of text in your document. - Reduce the value of [**Max response tokens**](https://autosheet.com/docs/gpt-for-docs/configure/configure-behavior#set-maximum-response-size). - Select a model with a higher [limit](https://autosheet.com/docs/resources/models/rate-limits). ## Is there a limit to the number of requests I can send? AI providers enforce rate limits to ensure smooth and fair access for everyone. [Learn more.](https://autosheet.com/docs/resources/models/rate-limits) ## Why are some features locked? Some features depend on your plan, not just your credit balance. With subscription pricing, features such as team management, centralized billing, and Bring Your Own AI (BYOAI/K/E) are only available on the Business and Enterprise plans. If you have credits available and you're still prompted to upgrade, the feature you're trying to use is not included in your current plan. For more information about what each plan includes, see the [plan comparison](https://autosheet.com/pricing). ## How do I uninstall the GPT for Docs add-on? The uninstallation process is described [here](https://autosheet.com/docs/gpt-for-docs/setup/uninstall). ## Which models can I use? Check our [list of supported models](https://autosheet.com/docs/resources/models/ai-providers-and-models). ## Which languages can I use? Models support many languages, but results vary by model, language, and task. Specify the language you want the response to use. If your prompt or source material contains several languages, make the instructions and expected output language clear. ## How do I get support? We recommend checking our [troubleshooting](https://autosheet.com/docs/gpt-for-docs/troubleshooting) and [FAQ](https://autosheet.com/docs/gpt-for-docs/faq) pages. If that doesn't help, [submit a support request](https://support.autosheet.com/hc/en-us/requests/new) and we'll get back to you. Our support hours are from Monday to Friday: - 7:30 PM - 12:00 PM (New York) - 1:30 AM - 6:00 PM (Paris) - 5:00 AM - 9:30 PM (India) - 8:30 AM - 1:00 AM (Tokyo)