GPT-6 Luna

OpenAI's lightweight reasoning model — $0.10 / $0.50 per million tokens, 20x cheaper than GPT-6 Sol.

GPT-6 Luna launched on September 22, 2026 alongside GPT-6 Sol. It brings reasoning capabilities to cost-sensitive workloads at 20× less than Sol while running 1.7× faster. This page collects its pricing, specs and every head-to-head comparison, plus a calculator for estimating your monthly token cost.

Last updated 2026-10-02 · Data from OpenAI and Artificial Analysis

GPT-6 Luna Key Specifications

SpecGPT-6 Luna
DeveloperOpenAI
API model IDgpt-6-luna
ReleasedSep 22, 2026
Context window1.05M tokens
Max output128K tokens
Input / output price$0.10 / $0.50 per MTok
Knowledge cutoffMay 18, 2026
ThinkingReasoning model, effort configurable
Default effortmedium
Input / outputText + images in, Text out
PlatformsOpenAI API, Azure OpenAI, ChatGPT
RetirementAt least 6 months notice per OpenAI policy

GPT-6 Luna Pricing

GPT-6 Luna costs $0.10 per million input tokens and $0.50 per million output tokens — making it one of the cheapest reasoning models available. Fast mode doubles those prices for faster output.

TierPrice per MTok
Standard API$0.10 input / $0.50 output
Batch API (50% off)$0.05 input / $0.25 output
Cache write$0.13
Cache read$0.01
Fast mode$0.20 / $1.00 per MTok

See Luna vs Sol for a detailed cost and quality comparison.

GPT-6 Luna Cost Calculator

Estimate your monthly GPT-6 Luna bill from your own request volume and token counts.

Estimated monthly cost (30 days)

$13.50

Average cost per request

$0.0004

Token-price arithmetic only, using GPT-6 Luna list prices. Cache writes, fast mode and tool or search surcharges are not included, and the Batch estimate ignores prompt caching. Real cost per task also depends on how many reasoning tokens the model uses, so measure on your own workload.

What GPT-6 Luna Is Best For

High-volume classification

Labeling, routing and categorization at massive scale for pennies.

Real-time agents

Latency-sensitive agents that need reasoning at very fast speeds.

Cost-first workloads

Use cases where budget is the primary constraint and quality is sufficient.

Rapid prototyping

Building and iterating on AI features without worrying about API bills.

GPT-6 Luna vs Other Models

GPT-6 Luna: FAQ

  • How much does GPT-6 Luna cost?

    GPT-6 Luna costs $0.10 per million input tokens and $0.50 per million output tokens — 20x cheaper than GPT-6 Sol. Batch requests are half price ($0.05 / $0.25), cache reads cost $0.01 per million tokens, and fast mode costs $0.20 / $1.00 per million tokens. A typical 50K-input, 10K-output request costs about $0.01.

  • Can I use GPT-6 Luna for free?

    GPT-6 Luna is available in ChatGPT for subscribers. API access requires an OpenAI account with credits. At $0.10/$0.50 per million tokens, costs are negligible for most use cases — processing 10 million input tokens costs just $1.

  • How fast is GPT-6 Luna?

    GPT-6 Luna is classified as very fast — 1.7x faster than GPT-6 Sol. This makes it ideal for latency-sensitive applications like real-time agents and high-volume classification tasks.

  • How does GPT-6 Luna compare to GPT-6 Sol?

    Luna is 20x cheaper and 1.7x faster. Sol leads all benchmarks, but Luna nearly matches on DeepSWE. For tasks that don't need Sol's quality ceiling, Luna offers dramatically better economics.

  • How does GPT-6 Luna compare to Claude Haiku 4.5?

    Both are lightweight reasoning models. Luna costs $0.10/$0.50 per MTok while Haiku 4.5 costs $1/$5 per MTok — Luna is 10x cheaper. Haiku 4.5 has a 200K context window vs Luna's 1.05M. The best choice depends on your quality and context requirements.

  • What is GPT-6 Luna's context window?

    GPT-6 Luna has a 1.05M-token context window and a maximum output of 128K tokens.

  • Is GPT-6 Luna good for coding?

    Luna is capable for everyday coding tasks and nearly matches Sol on the DeepSWE benchmark. For complex agentic coding workflows, Sol or Opus 5.5 will produce better results.

  • What is GPT-6 Luna's knowledge cutoff?

    The knowledge cutoff is May 18, 2026. OpenAI provides at least 6 months notice before deprecation per their policy.

  • Is GPT-6 Luna available on Azure?

    Yes. GPT-6 Luna is available through the OpenAI API, Azure OpenAI, and ChatGPT.

Sources & References