OpenAI has made its Decisions API generally available, enabling applications and agents to make near real-time decisions up to 10 times faster than using GPT-6 Luna through the Responses API. The API is designed to handle structured inputs and bounded answers, reducing latency and cost at scale.
TL;DR
- OpenAI's Decisions API is now generally available, offering faster, more cost-effective decision-making for applications.
- The API integrates with GPT-Live for real-time voice applications and supports up to 400 questions per request.
- Early adopters are using the Decisions API for various use cases, from categorizing transactions to enhancing language-learning experiences.
What happened
OpenAI has launched the generally available version of its Decisions API, built with GPT-6 Luna. The API focuses on making fast, structured decisions from text, images, and structured inputs, up to 10 times faster than GPT-6 Luna through the Responses API. Since its public beta, OpenAI has added several features, including native GPT-Live integration, prompt caching, and inference residency in the US and EU. Developers can now pass native JSON inputs, include images in question instructions, and ask up to 400 questions per request, double the previous limit. The API also supports three answer types: probability, choice, or score, making it versatile for various use cases.
Why it matters
The Decisions API allows developers to build more responsive applications by reducing latency and cost associated with decision-making processes. This is particularly useful for applications that require real-time interactions, such as customer support and language-learning experiences. The API's integration with GPT-Live enables natural, expressive conversations in voice applications, enhancing user engagement. For investors, this launch highlights OpenAI's commitment to expanding its product offerings and improving the functionality of its AI models. However, the effectiveness of the Decisions API will depend on its adoption by developers and the specific use cases it addresses.
Key facts
- The Decisions API is generally available as of the announcement.
- It is built with GPT-6 Luna and offers up to 10x faster decision-making compared to GPT-6 Luna through the Responses API.
- The API supports up to 400 questions per request, double the previous limit.
- It integrates natively with GPT-Live for real-time voice applications.
- The API costs $0.10 per million input tokens, with cached input costing $0.01 per million tokens.
- Regional processing is available in the US and EU, with Zero Data Retention available to eligible customers.
- Early adopters include OpenAI's own teams, using the API for various internal workflows.
Context
OpenAI's Decisions API is part of a broader trend in the AI industry to make AI models more efficient and cost-effective. As AI applications become more complex, the need for faster, more structured decision-making processes increases. OpenAI's focus on reducing latency and cost with the Decisions API aligns with this trend and positions the company as a leader in developing practical AI solutions. The API's integration with GPT-Live also highlights the growing importance of voice applications in the AI landscape, as companies seek to enhance user engagement through more natural, expressive interactions.