OpenAI launches Decisions API for fast, structured AI judgments
OpenAI released the Decisions API on October 22, 2026, designed for quick, structured AI decisions rather than long-form text. It returns predefined answers—probabilities, choices, or scores—in fixed formats like JSON, enabling apps to route inquiries, classify content, or trigger actions without waiting for full responses.
Key points
- OpenAI’s Decisions API returns structured answers (probabilities, choices, scores) in fixed formats like JSON for quick app decisions
- Officially 10x faster than GPT-6 Luna’s Responses API, with support for up to 400 simultaneous queries per request
- Integrates with GPT-Live for real-time voice interactions, but requires manual threshold-setting for confidence-based routing
The API is about 10 times faster than using GPT-6 Luna with the Responses API, according to OpenAI’s benchmarks. It supports up to 400 simultaneous queries and integrates with GPT-Live for real-time voice interactions, such as adjusting avatars or escalating issues. However, OpenAI emphasizes that confidence scores are not ‘automatic execution permits’ and warns developers to test thresholds with real-world data to avoid errors. The API is positioned as a complement—not replacement—for text generation or tool execution, with a focus on predefined workflows.
Model page: GPT-6 Luna →
The story so far
2 episodes →- OpenAI launches Decisions API for fast, structured AI judgmentsthis story
Does the screen move without waiting for the AI's reply? A gentle read of OpenAI's 'Decisions API'
note.com · 28 October 2026
Loading the full article…
This text was published by note.com and written by つるとーのひとりごと. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Agents & Tools
All →- Sierra adds compliance tools to Horizon Agent Platform · 1 src
- Supabase adds agent-friendly features for AI-driven app development · 1 src
- Anthropic launches Claude Duet, a proactive AI assistant for US paid users · 2 src
- AI agents manage fantasy soccer team but aren’t better than humans yet · 1 src
- Anthropic study finds AI agents sway experts’ decisions 86% of the time · 4 src
Comments
via GitHub Discussions