OpenAI has officially opened up API access to its deep research models, providing developers with a powerful set of tools including automatic web search, data analysis, MCP (Model Communication Protocol), and code execution.

The models now available include the deep research versions of o3 and o4-mini, which have already been used in ChatGPT. Now, developers can directly call these models via API. These models are particularly suitable for complex tasks that require accessing the latest information and performing advanced reasoning.

OpenAI

In terms of functionality expansion, models such as o3, o3-pro, and o4-mini now support web search functionality. OpenAI has also adjusted its pricing strategy: the starting price for reasoning web search service is $10 per 1,000 calls, while the web search price for GPT-4o and GPT-4.1 has been significantly reduced to $25 per 1,000 calls.

To enhance the developer experience, OpenAI has also introduced a webhook feature. This feature automatically sends notifications to developers when a task is completed, eliminating the need for manual continuous checks of the task status. OpenAI specifically recommends using webhooks for long-running tasks such as deep research to improve system reliability and development efficiency.

This series of updates marks an important advancement for OpenAI in the API services field, offering developers more powerful and cost-effective AI tool options.