GPT-5.6 Luna logo

MODELS

GPT-5.6 Luna

by OpenAI

modelopenaigpt-5.6-luna
Add to comparisonVisit WebsiteDocumentation

Overview

OpenAI’s fastest, lowest-cost GPT-5.6 model for cost-sensitive, high-volume text, coding, reasoning, and tool-use workloads via the OpenAI API.

Details

GPT-5.6 Luna is an OpenAI API model in the GPT-5.6 family. Supplied OpenAI sources describe it as the fastest, lowest-cost or most cost-efficient GPT-5.6 model and position it for cost-sensitive, high-volume workloads. The model-specific API page identifies GPT-5.6 Luna as a fast, high-reasoning model with text and image input, text output, a 1,050,000-token context window, 128,000 max output tokens, and Responses API tools. OpenAI’s GPT-5.6 guidance says to use gpt-5.6-luna for efficient high-volume workloads, and the preview system card covers safety evaluations, risk designations, and safeguards for Sol, Terra, and Luna.

When to Use

Use for cost-sensitive or high-volume API workloads where OpenAI’s GPT-5.6 family is appropriate. Use for text coding reasoning or tool-use tasks where speed and lower cost are priorities. Use when OpenAI’s API guidance recommends gpt-5.6-luna for efficient high-volume workloads.

Getting Started

  1. Review the GPT-5.6 Luna model documentation for supported inputs
  2. outputs
  3. context window
  4. max output tokens
  5. and Responses API tool support.
  6. Check OpenAI’s latest-model guidance for the current recommendation to use gpt-5.6-luna in efficient high-volume workloads.
  7. Review OpenAI API pricing for gpt-5.6-luna before production use.
  8. Review the GPT-5.6 preview system card and OpenAI safety-help guidance for safeguards that may affect biological or cybersecurity requests.
  9. Run a small evaluation on your own text
  10. coding
  11. reasoning
  12. or tool-use workload.

Key Features

  • •Described by OpenAI as the fastest GPT-5.6 model.
  • •Described by OpenAI as the lowest-cost or most cost-efficient GPT-5.6 model.
  • •Positioned for cost-sensitive
  • •high-volume API workloads.
  • •Supports text and image input with text output
  • •according to the model-specific API page.
  • •Model-specific API page lists a 1
  • •050
  • •000-token context window and 128
  • •000 max output tokens.
  • •Supports Responses API tools
  • •according to the model-specific API page.

Capabilities

  • •text-input
  • •image-input
  • •text-output
  • •coding
  • •reasoning
  • •tool-use
  • •responses-api-tools
  • •large-context

Last updated Jul 30, 2026