All Models
OpenAIDecisionProprietary Commercial APIReleased: 2026-09-29

OpenAI Decisions

OpenAI Decisions is OpenAI's dedicated high-throughput decision-making API powered by GPT-6 Luna, accessed via the POST /v1/decisions endpoint. Designed for agent routing, automated content moderation, and classification, it processes typed questions (predicates, categorical choices, and rubric scores) roughly 10x faster than standard generation, billed at $0.10 per million input tokens with zero output-token fees.

Foundation DNA96/100
HYBRID-REASONING ArchProprietary (GPT-6 Luna Decision Engine) params
At a Glance Specifications
Context Window

128k

~96k words input

Token Pricing

$0.1 / $0

Prompt / Completion per 1M

Top Benchmark

82 ms

Decision Latency

Hardware Tier

Cloud Managed

Zero local VRAM

6-Axis Foundation Model DNA

Holistic Capability Distribution

Normalized 0-100 rating vectors balanced across intelligence, throughput, and economics.

Overall Index:96/100
ReasoningCodingSpeedEconomicsContextFlexibility
Context: 92/100 (Window Horizon)
ReasoningGPQA / Logic / Math
95/100
CodingSWE-bench / LCB
90/100
SpeedTokens / Second
99/100
EconomicsCost Efficiency
99/100
ContextWindow Horizon
92/100
FlexibilityTool Use & Open
94/100
Performance Evaluations

Verified LLM Benchmarks

Empirical evaluations across academic reasoning, software engineering, agentic execution, and human preference.

Decision Latency Headline
82ms
Production Endpoint Latency Harness
Agentic ExecutionOpenAI DevDay 2026 Technical Presentation
Predicate Verification Accuracy Headline
98.2% accuracy
Binary Condition Test Suite
Academic ReasoningOpenAI Decisions Technical Evaluation
Multimodal Classification Headline
95.8% accuracy
Vision + Text Inspection Suite
Academic ReasoningOpenAI Decisions Multimodal Eval
Agent Routing Precision
97.4% precision
Autonomous Loop Triage
Agentic ExecutionOpenAI Agent Workflows Benchmark
Cost Reduction vs. Standard Chat
92% savings
Classification Cost Comparison
CodingOpenAI DevDay Pricing Analysis
Official Release Visuals & Benchmark Artifacts

OpenAI Decisions Empirical Results & Generational Scaling

Verified release visuals published by Inception Labs comparing Mercury 3 against GPT-6 Luna and Claude Haiku 5.5.

Decisions API vs. Standard LLMs: Workflow Accuracy & Cost
Decision Correctness & Schema Reliability Comparisons
Real-Time Provider Telemetry

Inference Latency & Generation Throughput

Empirical telemetry benchmarking generation throughput (Tokens/Second) and Time to First Token (TTFT).

Peak Generation Speed
1800tokens/sec

Measured under concurrent production load

Lowest Time to First Token
75ms

p95 latency: 160 ms

Telemetry Verification
99.98%uptime

Across 50,000 monitored requests

Production Trade-Offs

Architectural Strengths vs. Operational Considerations

An objective technical balance sheet analyzing deployment advantages and production constraints.

Operational Advantages

Key Architectural Strengths

  • Native multimodal context evaluation: inspects text and image documents together against typed question schemas.
  • Powered by GPT-6 Luna world intelligence with specialized non-autoregressive decision classification heads.
  • Zero output token billing: only input tokens are billed ($0.10 / 1M), yielding up to 92% cost reduction compared to chat completions.
  • Operating up to 10x faster than standard generation (~80ms latency) for high-velocity agent execution loops.
  • Calibrated probabilities for predicates (0.0 to 1.0) and normalized categorical distributions.
Production Constraints

Operational Considerations

  • Accessible exclusively via POST /v1/decisions; does not generate free-form text or conversational replies.
  • Requires explicit structured schema definitions for each question (predicate, choice, or score).
Technical Specifications

Hardware & Execution Parameters

Architecture Typehybrid-reasoning
Total ParametersProprietary (GPT-6 Luna Decision Engine)
Attention MechanismSparse Speculative Cross-Attention
Context Window128,000 tokens
0
Supported Modalities
textimagecodejson
License Typeproprietary
Commercial Use Allowed
Model DistillationProhibited
Zero Data RetentionAvailable
Knowledge Cutoff2026-09
Model Heritage & Evolutionary Lineage

Genealogical Provenance

Dedicated fast decision interface built on GPT-6 Luna, announced at OpenAI DevDay 2026.

Base Precursor

openai gpt 6 luna

Current Architecture

OpenAI Decisions

openai gpt 6 lunaopenai gpt 6 solopenai gpt 6 astra
Engineering Deep-Dive

Architecture Engineering & Capability Deep-Dive

Topology & Attention Mechanics

The Decisions API leverages GPT-6 Luna with a specialized decision-head topology that bypasses autoregressive token emission. When a request is received at /v1/decisions, the input context is encoded once, and all specified questions (predicates, choices, and scores) are evaluated concurrently against the final layer representations.

Evaluation Profile & Reasoning Capabilities

Combines GPT-6 Luna's world knowledge with calibrated softmax outputs, yielding precise probabilities between 0.0 and 1.0 for predicates, normalized choice distributions, and weighted rubric ratings.

Inference Economics & Cost Optimization

Eliminates token output charges entirely. Processing 1,000 tickets with a 500-token input costs just $0.05 total.

Primary Production Workloads & Suitability

Agentic loop orchestration and tool invocation gating
Multimodal image inspection, visual damage assessment, and receipts parsing
High-speed content moderation, PII scanning, and safety evaluation
Dynamic search query routing between retrieval-augmented generation and vector databases
Compute & Deployment Architecture

Cloud Deployment Architecture & Hardware Sizing

Cloud Deployment Architecture

Distributed enterprise serving infrastructure, autoscaling semantics, and runtime execution profile.

Cloud APIZero Local VRAM Required

Proprietary commercial model hosted on managed multi-region clusters with automatic autoscaling, high availability, and prompt prefix caching.

SLA: 99.9% Cloud Availability
Rate Limit: Scaled via Provider Quota
Cache: Automatic Prefix Compression
Serving Model

Multi-Tenant Cloud API

Continuous autoscaling with regional failover

Hardware Overhead

0 GB Local VRAM

Inference executed on hyperscaler accelerator clusters

Context Capacity

128k tokens

Supported with automated KV cache prefix compression

SDK Installation

$npm install openai
$pip install openai
Inference Economics & Total Cost of Ownership

Commercial Token Rates & Cost Estimator

Base Input Tokens

$0.10 / 1M

Standard prompt context ingestion

Base Output Tokens

$0.00 / 1M

Generated completion & thinking tokens

Prompt Cache ReadUp to 80% Off

$0.03 / 1M

Re-used prompt prefix context

Monthly Production Cost Simulator

Daily Prompt Ingestion5M tokens/day (~150M/mo)
Daily Completion Generation1M tokens/day (~30M/mo)
Prompt Prefix Cache Hit Rate40%
Estimated Monthly Spend

$10.50

Includes prompt caching

Monthly Cache Savings

-$4.50

Saved via prefix context caching

Frontier Alternatives

Comparable Foundation Architectures

Alternative foundation models with comparable intelligence benchmarks, context windows, and deployment profiles.

TypeSafe AIDecision

Jev AI

Jev is TypeSafe AI's breakthrough non-autoregressive System One decision model. Rather than generating conversational prose, Jev evaluates state payloads (JSON or text) against explicit typed questions—including binary condition probabilities (noul), categorical choices, and multi-tier rubric scores—in parallel, returning calibrated probabilistic judgments at $0.042 per million input tokens with zero output token fees.

Context:65.536k ctx
Parameters:Proprietary (Discriminative Decision Head)
Input Rate:$0.042/1M
Compare Architecture
InceptionGeneral

Mercury 3

Mercury 3 is Inception's frontier diffusion large language model (dLLM) co-founded by Stanford Professor Stefano Ermon, delivering an unprecedented 1,068 tokens per second inference speed via parallel denoising across a native 1-million-token context window.

Context:1,000k ctx
Parameters:Proprietary (~40x Mercury 1)
Input Rate:$0.1/1M
Compare Architecture
OpenAISmall / On-Device

GPT-6 Luna

Ultra-fast low-latency frontier model in the GPT-6 generation, streaming at 130+ tps across a 1-million-token context window at $0.20/$0.80 economics.

Context:1,000k ctx
Parameters:Proprietary
Input Rate:$0.2/1M
Compare Architecture
OpenAIGeneral

GPT-3.5 Turbo

OpenAI's foundational conversational model that launched the ChatGPT era, optimized for fast dialogue, summary generation, and lightweight JSON tasks.

Context:16.385k ctx
Parameters:Proprietary
Input Rate:$0.5/1M
Compare Architecture
Developer Quickstart

SDK Installation & API Integration

$npm install openai
$pip install openai
import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.OPENAI_API_KEY,
});

// Calling the dedicated /v1/decisions endpoint
const response = await fetch("https://api.openai.com/v1/decisions", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.OPENAI_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "gpt-6-luna",
    input: "Customer message: 'I love your product, but the checkout page failed twice.'",
    questions: [
      {
        type: "predicate",
        name: "is_complaint",
        instructions: "Is this message reporting an error or expressing frustration?"
      },
      {
        type: "choice",
        name: "category",
        choices: ["billing_checkout", "product_feedback", "account_access"]
      },
      {
        type: "score",
        name: "churn_risk",
        scale: [1, 5],
        instructions: "Assess churn risk from 1 (unlikely) to 5 (imminent)"
      }
    ]
  }),
});

const data = await response.json();
console.log(data.answers);
Knowledge Base

Frequently Asked Questions about OpenAI Decisions

Engineering specifications, VRAM sizing, commercial licensing terms, token economics, and empirical benchmark evaluations.

Topic:

Announced at OpenAI DevDay 2026 and released in public beta in October 2026, the Decisions API (POST /v1/decisions) is an OpenAI interface powered by GPT-6 Luna that returns typed, structured judgments rather than free-form text. It runs ~10x faster than standard chat completions.

Verified Engineering FAQ

OpenAI Decisions supports three question types: 'predicate' (evaluates a condition and returns a probability from 0.0 to 1.0), 'choice' (selects an option from a developer-defined set of values), and 'score' (evaluates an input against an ordered rubric scale).

Verified Engineering FAQ
Machinoai Logo
Machinoai

Your edge in the AI era. Deep dives into machine learning concepts, paper explainers, verified industry jobs, and developer tools.

© 2026 Machinoai. All rights reserved.