Skip to main content

OpenAI Decisions API Hits Public Beta on GPT-6 Luna

OpenAI launches Decisions API in public beta on GPT-6 Luna, offering 10x faster typed answers at $0.10 per 1M input tokens with zero output charges.

AI-written
Inewgen
10 Oct 2026Source: MarkTechPost2 min read (0 views)
Share
OpenAI Decisions API Hits Public Beta on GPT-6 Luna

Stock photo for illustration only, not from the actual event

Font size
  • OpenAI launches Decisions API in public beta on the GPT-6 Luna model
  • Delivers processing speeds up to 10x faster than the standard Responses API
  • Returns structured typed probabilities, choices, and scores
  • Priced at $0.10 per 1 million input tokens with no output charges

OpenAI has officially announced the public beta rollout of its new Decisions API, operating on the GPT-6 Luna model. This release is engineered to handle complex data processing and automated decision-making tasks with enhanced efficiency for developers worldwide.

A standout feature of this new interface is its remarkable performance boost. According to the release details, the API delivers typed answers—including probabilities, specific choices, and evaluation scores—approximately 10 times faster than the conventional Responses API.

software developer office desk workspace notebook computer

Stock photo for illustration only, not from the actual event

10xFaster than Responses API
$0.10Per 1M Input Tokens

Regarding the pricing structure, OpenAI has established a rate of $0.10 per 1 million input tokens for the Decisions API. Notably, the company has structured the service to incur no output charges whatsoever, allowing developers to scale their applications without worrying about additional token generation costs.

The introduction of the Decisions API on GPT-6 Luna highlights OpenAI's strategic shift toward supporting enterprise applications that require highly structured and rapid automated decisions. Eliminating output fees serves as a strong incentive for businesses processing high-volume analytical workloads.

Source: MarkTechPost

Comments

Leave a Comment
0/2000

Found something wrong in this article? Report an issue with this article