OpenAI Decisions API Hits Public Beta With 10x Faster Typed Answers

What changed

OpenAI launched the Decisions API on public beta with GPT-6 Luna. This new API delivers typed outputs including probabilities, choices, and scores around 10 times faster than the existing Responses API. Pricing is set at $0.10 per 1 million input tokens, with no charges for output tokens, making it a cost-effective alternative for applications focusing heavily on input-driven decision queries.

Why builders should care

The Decisions API speed and output format shift the game for developers building apps that rely on fast, structured decision-making from AI models. Rather than receiving a plain-text answer, users get granular typed data on likelihoods and decision scores, enabling tighter integration with automated workflows and decision pipelines. The faster response time means systems can scale more efficiently and reduce latency in real-time environments.

The practical takeaway

For operators running AI-driven pipelines, this upgrade lowers overhead by cutting computation time and output generation fees. Builders can improve throughput and enable richer decision analytics without ballooning costs. It also makes sense for use cases like A/B testing, risk assessments, and recommendation engines where probabilistic insight and speed directly impact outcomes and user experience.

What to watch next

Keep an eye on adoption patterns and performance reports from early users to see if the API lives up to its 10x speed claims in real-world settings. Also watch for any limitations or quirks in the typed output format that might require changes in client-side processing or orchestration logic. Pricing pressure on competing decision-oriented AI services could follow if OpenAI keeps costs low and performance high.

AI Quick Briefs Editorial Desk