AI BriefingOpenAI Decisions API generally available with GPT-6 Luna for 10x faster bounded decisions
·2026.10.23 04:00
Key point
Built with GPT-6 Luna, the API processes text, images, and JSON inputs to return probabilities, choices, or scores, now supporting up to 400 questions per request and native GPT-Live integration.
Details
- OpenAI has made the Decisions API generally available, built with GPT-6 Luna to handle questions with bounded answers up to 10x faster than the standard Responses API.
- The API supports three answer types: predicate (probability 0-1), choice (predefined options with confidence), and score (probability-weighted average of rubric levels).
- Pricing is $0.10 per million input tokens with no output-token charges; cached input costs $0.01 per million tokens.
- New features include native GPT-Live integration for real-time voice experiences, prompt caching, and inference residency in the US and EU.
- Developers can now pass up to 400 questions per request, double the previous limit, and use native JSON inputs or images in question instructions.
- Early adopters like Box report ~5x faster route selection with the same accuracy as the Responses API for document classification tasks.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
OpenClaw creator estimates $10 million token spend last month
CopilotKit Releases OpenIntelligentUI, an Open-Source Alternative to ChatGPT Intelligent UI