AI models

Models are the engines that power GPT for Work. Which model is the right one depends on your specific needs in terms of response quality, speed, privacy, and cost. GPT for Work supports models from Anthropic, Azure, DeepSeek, Google, Mistral, OpenAI, OpenRouter, Perplexity, and xAI. GPT for Work also supports open-source models through Ollama and any OpenAI-compatible API endpoint.

note

As a space owner or admin (Microsoft accounts only), you can control which models are available to users in your space through the Models section of the dashboard.

The following table summarizes the different kinds of models available in GPT for Work from zero set up to more advanced options:

Use	Setup	When	Available for
Models without an API key	None. Use the default model or select a different model.	You want to get started quickly without setting up API keys. ✅ Available models	Excel, Sheets, Word, Docs
Models with an API key 🔑	AI provider models: Set an API key for an AI provider.	You want to benefit from the terms of service you have with an AI provider, such as cost, privacy, and rate limits. You want to use a model that is not available without an API key. You want more control. ✅ Available models	Excel, Sheets, Word, Docs
	OpenRouter models: Set an API key for OpenRouter.	You want to access models from different providers without managing multiple API keys. You want to use models that are only available through OpenRouter. ✅ Available models	Excel, Sheets, Word
	OpenAI Assistants: Create an OpenAI Assistant and set an OpenAI API key.	You need to add custom capabilities to a model. You want to upload your own files as knowledge for the AI.	Excel, Sheets
Models from cloud endpoints	Azure OpenAI models: Set up an Azure OpenAI endpoint.	Your company is already using Azure with a consumption commitment. Using Azure OpenAI counts towards that commitment. You want to use models that are available through Azure OpenAI. ✅ Available models	Excel, Word
	OpenAI-compatible models: Set up an OpenAI-compatible platform endpoint.	You want to use models available on a cloud-based LLM platform such as Anyscale or Fireworks AI. You need a fast and powerful inference infrastructure that scales to your performance needs.	Excel, Word
Local models	OpenAI-compatible models: Set up an OpenAI-compatible server endpoint.	You want to use models available on a local LLM server such as LM Studio. You want to run models on a local machine. You do not want to send or receive any prompt or response data over the internet.	Excel, Word
	Ollama models: Set up an Ollama endpoint.	You want to use models available through Ollama. You want to run models on a local machine. You do not want to send or receive any prompt or response data over the internet. ✅ Available models	Excel, Word

What's next

Get started with the option that works best for you:

• Select a model.

• Set and manage your API keys.

• Use OpenRouter models.

• Use OpenAI Assistants.

• Use OpenAI-compatible models through API endpoints.

• Use Ollama models.