New guide: Using GPT-6 Astra — how to use GPT-6 Astra for reasoning, coding, professional work, and tool-using workflows, plus migration from GPT-5.6.
Changelog
Models
New guide: Using GPT-6 Astra — how to use GPT-6 Astra for reasoning, coding, professional work, and tool-using workflows, plus migration from GPT-5.6.
Model ID docs now include gpt-6.
Tools and Responses API
New guide: Async tool calling — start asynchronous tool calls, deliver results later, and let the model wait for specific results when needed.
New guide: Mid-turn steering — steer a Responses API workflow over WebSockets, follow continuations, and handle queued input and tool results.
New guide: Computer use integration recipes — environment setup, action handlers, custom execution services, and consent guidance.
Updated description for Computer use to cover connecting a code-execution harness or computer tool for browser and desktop interfaces through the Responses API.
Safety checks is now titled Safety classifiers, with an updated focus on how API safety classifiers identify potential abuse and how safety identifiers help limit disruption to other users.
Cybersecurity now documents API cybersecurity safeguards, authorized access, request-level checks, access restrictions, and appeals.
Under-18 guidance (formerly “Under 18 API Guidance”) now covers regulatory standards, safety standards, and best practices for API apps that serve users under 18.
