Supported AI models
Supported AI models
Box supports a variety of AI models, categorized along two dimensions: access level and capability tier.
Access Levels
Core Models
These models are built into Box AI and available by default for all customers. No configuration is required.
Customer-Enabled Models
These models require activation by Box admins in the Admin Console or a request to Box to enable them. Some models may be subject to additional terms or pricing.
Capability Tiers
Standard Models
Designed for high-speed, cost-efficient tasks like basic summarization, Q&A, and structured data extraction from shorter or simpler documents. Ideal for high-volume, low-complexity use cases.
Premium Models
Offer more advanced reasoning, larger context windows, and better performance on long-form, complex, or domain-specific content. Suitable for sophisticated tasks like multi-step reasoning, understanding large taxonomies, and analyzing lengthy or unstructured documents.
Using models
How to use the supported AI models:
- get the default AI agent configuration,
- override the AI agent configuration used in
POST 2.0/ai/ask
,POST 2.0/ai/text_gen
,POST 2.0/ai/extract
,POST 2.0/ai/extract_structured
endpoints.
When using the model
parameter your API calls, use the API Name visible on each tile and model card.
For example, to get the AI agent configuration for a specific model, use the model parameter and provide the azure__openai__gpt_4o_mini
API name. Make sure you use two underscores after the provider name.
Core Box AI Models
Box AI is powered by the following AI models. These models are integrated with Box AI to facilitate various use cases while adhering to enterprise grade standards. Below, you’ll find information about each model, including its capabilities, intended applications, and applicable usage guidelines.
Customer-enabled models
Certain Box AI customers may enable additional AI models upon their request and/or otherwise made available to them through their admin console. Use of these models may be subject to additional terms. By selecting a customer-enabled model, customer acknowledges that their data may be processed by additional subprocessors of their choice.