OpenAI has made its Decisions API available to all customers, introducing a faster and cheaper way for developers to use artificial intelligence for tasks that require quick decisions.
The company announced the wider release on Thursday, October 22, after initially introducing the technology at its developer event on September 29.
According to OpenAI, the Decisions API can make decisions up to 10 times faster than its underlying model through the traditional Responses API, making it suitable for applications that require near-instant responses.
How the Decisions API Works
Unlike traditional large language models (LLMs), which generate detailed responses, the Decisions API focuses on selecting answers from predefined options.
Developers can provide questions and possible answers, allowing the model to classify information, choose actions, or calculate confidence scores.
The API supports three types of responses: probability estimates, choices between options, and scores based on predefined criteria.
OpenAI developed the technology by modifying its existing language model to reduce processing time and costs rather than training a completely separate model.
Nikunj Handa, OpenAI’s API product lead, told TechCrunch that the company reduced response times from one to three seconds to approximately 100–180 milliseconds.
Performance and Pricing
OpenAI has introduced the Decisions API with a focus on faster response times and lower processing costs.
According to Nikunj Handa, OpenAI’s API product lead, the company reduced decision-making time from one to three seconds to approximately 100–180 milliseconds.
| Feature | Details |
|---|---|
| Traditional AI response time | 1–3 seconds |
| Decisions API response time | 100–180 milliseconds |
| Input pricing | $0.10 per million tokens |
| Cached input pricing | $0.01 per million tokens |
| Output token charges | None |
| Maximum questions per request | 400 |
| Supported inputs | Text, images, and structured data |
Early customers include Box, Datadog, Ramp, and Speak.
Speak CTO Andrew Hsu said the language-learning platform reduced the time between AI actions from 1.4 seconds to 0.13 seconds, representing an improvement of more than 90%.
Competition With TypeSafe
OpenAI’s release brings it into competition with TypeSafe, the startup behind the Jev decision-making model.
TypeSafe recently reached a valuation of approximately $7.6 billion. Other companies, including Liquid and Base Lab, have also introduced competing models.
The following table compares their positions on two public benchmarks, based on the reported rankings.
| AI Model | AIM System One Index | Artificial Analysis Decision Index |
|---|---|---|
| TypeSafe Jev | 7th | 3rd |
| OpenAI Decision Model | 8th | 14th |
| Liquid D1 Ultra | Strong performer | Strong performer |
| Base Lab Elephant-1 | Strong performer | Strong performer |
Exact rankings for Liquid D1 Ultra and Base Lab Elephant-1 were not supplied, so they are not assigned numerical positions. Rankings can also change as benchmarks are updated.
AI Models Could Become Even Faster
OpenAI believes further improvements could reduce response times to approximately 60 milliseconds.
Handa suggested that such speeds could eventually allow developers to make AI decisions for individual frames in videos.
Meanwhile, TypeSafe has expanded Jev with image support and conditional decision-making features.
The growing competition highlights increasing demand for AI systems that can handle repetitive decisions without the higher costs and processing times of traditional language models.
OpenAI expects faster and cheaper decision models to support more applications, including real-time voice assistants, image classification, automated workflows, and interactive software.
The post OpenAI Launches New AI Tool That Makes Decisions Up to 10 Times Faster appeared first on ProPakistani.