Skip to main content

Select a model

GPT for Work supports a range of models that you can use for your AI work. The default model is OpenAI gpt-4o. You can select a different model if you want to use or explore other options. To help you find the model that offers the best balance of accuracy, speed, and cost for your use case, see our recommended models.

The available models depend on:

Select a model without an API key

  1. Open a Microsoft Excel workbook.

  2. In the Home tab, click GPT for Excel Word.

    GPT for Excel Word in the Home ribbon

    info

    You can also select Home > Add-ins > My Add-ins > GPT for Excel Word.

  3. Expand the model switcher.

  4. Select a model.

    Models that are available without any setup are listed under No API key needed.

GPT for Excel uses the selected model to generate all responses.

info

Select a model with an API key

Prerequisites
  • You have set an API key for the AI provider whose model you want to use.

  • In Excel and Word, if you want to use an Azure model or an OpenAI-compatible model from a custom endpoint, you have set up the correct endpoint.

  1. Open a Microsoft Excel workbook.

  2. In the Home tab, click GPT for Excel Word.

    GPT for Excel Word in the Home ribbon

    info

    You can also select Home > Add-ins > My Add-ins > GPT for Excel Word.

  3. Expand the model switcher.

  4. Select a model with a key icon.

    Models that are available with an API key are listed under Use your API key or endpoint and have a key icon in front of their name.

GPT for Excel uses the selected model to generate all responses.

info

As a general rule, OpenAI models are more reliable and less likely to decline prompts than models from other providers.

The following table lists our recommended models available both without and with an API key. For more information about which add-on supports which models, see AI providers & models.

ModelProviderDescription

gpt-4o

OpenAI

Best overall balance of easy prompting, speed, quality, reliability, and cost. We recommend this model for most use cases.

gpt-4.1-nano

OpenAI

Fast, cost-effective model with strong reasoning and language capabilities. Follows instructions closely.

o4-mini

OpenAI

Fast, cost-effective reasoning. Great for analysis tasks.

claude-3.5-sonnet

Anthropic

Great at content creation, summarization, and nuanced reasoning across a wide range of tasks.

gemini-2.5-flash

Google

Cost-effective model ideal for lightweight tasks and rapid responses. Supports reasoning and web search (enabled from the model settings).

sonar

Perplexity

Cost-effective search model optimized for simple queries. Grounded in recent web search data.

sonar-pro

Perplexity

Advanced search model designed for complex queries. Grounded in recent web search data.

To learn more about a model, hover over the model name in the model switcher. A tooltip opens with detailed information about the model.

Model tooltip in the model switcher

What's next

Learn more about models:

Put your selected model to use:

  • Use bulk AI tools to run prompts on entire spreadsheet columns at once without writing any formulas.

  • Use GPT functions to run prompts from inside spreadsheet cells.