--- title: "OpenAI Launches New Decisions API: Instant AI Challenges Startups" description: "OpenAI has widely released the Decisions API, reducing AI request processing times to 100-180 milliseconds and challenging startup TypeSafe." date: 2026-10-23T08:16:21.388Z lang: en url: https://xab.info/en/posts/openai-launches-new-decisions-api tags: [openai, decisions-api, artificial-intelligence, typesafe, chatgpt-luna, tech-news] author: "Alexander Buran" publisher: "XAB.info" --- # OpenAI Launches New Decisions API: Instant AI Challenges Startups ![Transparent futuristic desktop monitor displaying OpenAI API code and developer interface](https://xab.info/media/2026/10/23/openai-otkryla-noviy-decisions-api/openai-otkryla-noviy-decisions-api-1.webp) ## Key Points - OpenAI launched broad access to Decisions API for fast AI decision-making. - Request processing time reduced from 1-3 seconds to 100-180 milliseconds. - The tool is built on a reconfigured ChatGPT Luna model. - Early tests at Datadog, Ramp, and Speak showed significant performance gains. OpenAI has announced broad access to the Decisions API, a new specialized tool designed for ultra-fast and cost-effective AI decision-making. The innovation rollout comes as the tech giant seeks to capture a leading position in a promising segment previously dominated by niche startup TypeSafe with its Jev system. Processing time for the new product has been successfully reduced from the traditional 1-3 seconds down to a record-breaking 100-180 milliseconds. ### Technological Features and Performance Speed Decision-making models allow artificial intelligence to swiftly choose from available options and assess its own confidence level in the result. Instead of building an entirely new architecture from scratch, OpenAI's development team reconfigured the existing ChatGPT Luna model, optimizing it for rapid request processing and reduced operating costs. Nikunj Handa, Head of API Product at OpenAI, noted that the company's engineers closely studied the market success of similar solutions and swiftly created their own competitive version. The primary technological breakthrough of the Decisions API is its response speed, dropping to 100-180 milliseconds. In the long term, developers plan to introduce specialized training that will reduce latency down to 60 milliseconds. Such a metric will open up opportunities for the full processing of every single frame in a video stream in real time. ### Initial Deployment Results and Competitive Landscape During the preliminary testing phase, companies such as Datadog, Ramp, and Speak utilized the new interface to replace previous OpenAI tools. Notably, the popular tutoring service Speak managed to cut the delay between user actions from 1.4 seconds down to 0.13 seconds, while also improving the overall sensitivity of the Dots voice assistant. Nevertheless, the rapid decision-making solutions market remains highly competitive. TypeSafe CEO Diogo Almeida expressed skepticism regarding standard benchmarks, recommending that businesses evaluate models on real-world operational scenarios. In public indices, developments from Liquid and Base Lab remain leaders, while the updated Jev system retains popularity on the OpenRouter platform and ranks high in the specialized JevMan benchmark. ### Future Prospects and Market Impact The emergence of accessible tools with minimal response times marks the industry's transition to fundamentally new artificial intelligence usage scenarios. Previously, complex automation tasks were hindered by high latency and significant computational costs, but model optimization erases these barriers. The combination of high speed, multimodality, and reduced costs enables businesses to integrate AI into voice interfaces, interactive support systems, and next-generation multimedia applications. ## Fact-Check & Sources - [OpenAI открыла новый API: ИИ может принимать решение почти мгновенно](https://www.rbc.ua/ukr/news/openai-vidkrila-noviy-api-shi-mozhe-uhvalyuvati-1792742099.html) - The primary source confirms the launch of Decisions API, latency figures of 100-180 ms, and the involvement of Speak, Datadog, and Ramp. - [OpenAI научила ИИ говорить и слушать одновременно — полнодуплексная модель GPT-Live-1 доступна ...](https://3dnews.ru/1148357/openai-nauchila-ii-govorit-i-slushat-odnovremenno-polnodupleksnaya-model-gptlive1-dostupna-razrabotchikam) - Подтверждено по источнику 3dnews.ru - [OpenAI открыла API голосовой модели GPT-Live-1 для разработчиков сторонних приложений](https://www.pravda.ru/news/technolgies/2406512-openai-gpt-live-api-launch/) - Подтверждено по источнику pravda.ru - [OpenAI открыла доступ к GPT-6 для всех пользователей](https://www.sostav.ru/publication/openai-otkryla-dostup-k-gpt-6-dlya-vsekh-polzovatelej-87578.html) - Подтверждено по источнику sostav.ru ## FAQ ### What is OpenAI's new Decisions API? It is a tool for fast and cheap AI decision-making that allows choosing from options and assessing confidence within 100-180 milliseconds. ### Which model forms the basis of the new interface? Instead of creating a calibrated system from scratch, OpenAI engineers reconfigured the ChatGPT Luna model to optimize speed and reduce costs. ### Which companies were first to test the Decisions API? Preliminary testing involved Datadog, Ramp, and the tutoring service Speak, which reduced latency down to 0.13 seconds.