Decision Model Tracker

Last updated 16 Nov 2026. Sources last checked 16 Nov 2026.

All models · First seen 6 Oct 2026

Decisions API (GPT-6 Luna)

OpenAI · Released 6 Oct 2026

Get the APIDiscuss on X

A dedicated Decisions API that evaluates text, images, and structured inputs and returns a probability, a choice from a fixed set, or a rubric score. Aimed at classification, routing, and near real-time application decisions, and later integrated with GPT-Live.

Maker
OpenAI
Release date
6 Oct 2026
First seen
6 Oct 2026
Access
OpenAI Decisions endpoint with model gpt-6-luna. Public beta launched 6 Oct 2026; generally available 22 Oct 2026. By GA, the question limit rose to 400 per request and prompt caching, native JSON input, and GPT-Live integration were added.
Pricing (as publicly stated)
As stated in the 22 Oct 2026 general-availability post: $0.10 per million input tokens; cached input $0.01 per million tokens; no output-token charges. See OpenAI's pricing guide for cache-write rates and regional options.
Modalities
text, images, JSON
Question types
predicate, choice, score
Latency (as stated)
OpenAI said the API helps applications make near real-time decisions up to 10x faster than GPT-6 Luna through the Responses API.
Official announcement
https://openai.com/index/decisions-api/

Benchmark claims

These are claims as published by the maker or in the linked official post. This catalog has not independently rerun every benchmark.

OpenAI published customer-reported figures in the GA post, including Box's report of about 5x faster route selection on 30 business-document routing cases at the same accuracy as the Responses API, and Datadog's report of 40x faster p95 with comparable accuracy versus available frontier models. These are customer statements quoted by OpenAI, not a public third-party leaderboard.

Source: https://openai.com/index/decisions-api/

First public appearance is recorded as 6 Oct 2026 in OpenAI's API changelog (beta): https://developers.openai.com/api/docs/changelog. The linked announcement is the 22 Oct 2026 general-availability post.

What enterprises should know

Cost and pricing

As stated in the 22 Oct 2026 general-availability post: $0.10 per million input tokens; cached input $0.01 per million tokens; no output-token charges. See OpenAI's pricing guide for cache-write rates and regional options. Figures are what the maker stated in the linked source and may have changed.

Source: https://openai.com/index/decisions-api/

Lock-in

The official announcement describes access as follows: OpenAI Decisions endpoint with model gpt-6-luna. Public beta launched 6 Oct 2026; generally available 22 Oct 2026. By GA, the question limit rose to 400 per request and prompt caching, native JSON input, and GPT-Live integration were added. The announcement did not state a documented export path or switching process beyond that description.

Source: https://openai.com/index/decisions-api/

Security and privacy

The official announcement we sourced did not state a specific security certification or independent privacy audit for this product. If you use a hosted endpoint named in the announcement, inputs are sent to that host under the host's own terms.

Source: https://openai.com/index/decisions-api/

Maturity and availability

The maker's official announcement is dated 2026-10-06. Access was described as: OpenAI Decisions endpoint with model gpt-6-luna. Public beta launched 6 Oct 2026; generally available 22 Oct 2026. By GA, the question limit rose to 400 per request and prompt caching, native JSON input, and GPT-Live integration were added. OpenAI said the API helps applications make near real-time decisions up to 10x faster than GPT-6 Luna through the Responses API.

Source: https://openai.com/index/decisions-api/

Compliance notes

The official announcement we sourced did not state a specific regulatory approval or compliance certification for this product.

Source: https://openai.com/index/decisions-api/

What consumers should know

Cost and pricing

The announcement describes developer or enterprise access, not a consumer retail price. As stated in the 22 Oct 2026 general-availability post: $0.10 per million input tokens; cached input $0.01 per million tokens; no output-token charges. See OpenAI's pricing guide for cache-write rates and regional options.

Source: https://openai.com/index/decisions-api/

Lock-in

The announcement describes access as: OpenAI Decisions endpoint with model gpt-6-luna. Public beta launched 6 Oct 2026; generally available 22 Oct 2026. By GA, the question limit rose to 400 per request and prompt caching, native JSON input, and GPT-Live integration were added. Using a hosted API typically requires an account with the maker and is subject to that maker's terms.

Source: https://openai.com/index/decisions-api/

Security and privacy

Using a hosted endpoint means prompts and other inputs are sent to the maker or host named in the announcement. The announcement we sourced did not state a privacy policy unique to this product beyond that company's existing terms.

Source: https://openai.com/index/decisions-api/

Maturity and availability

The official announcement is dated 2026-10-06. Availability was described as: OpenAI Decisions endpoint with model gpt-6-luna. Public beta launched 6 Oct 2026; generally available 22 Oct 2026. By GA, the question limit rose to 400 per request and prompt caching, native JSON input, and GPT-Live integration were added. OpenAI said the API helps applications make near real-time decisions up to 10x faster than GPT-6 Luna through the Responses API. This summary does not say whether the product is suitable for a particular consumer task.

Source: https://openai.com/index/decisions-api/

Compliance notes

The official announcement we sourced did not state consumer-protection certifications specific to this product.

Source: https://openai.com/index/decisions-api/

Figures are as stated in the linked source on the date shown. Sources last checked 16 Nov 2026. Names and marks belong to their owners. This page is an independent summary, not a product page from OpenAI.