GPT-6 Luna AI features, pricing and GDPR (October 2026)

GPT-6 Luna is the most efficient model in OpenAI's GPT-6 family, meant for high-volume tasks such as classification, extraction and short answers where cost per request matters more than maximum capability.

Luna takes text and images and returns text, with a context window of 1,050,000 tokens, up to 128,000 output tokens and a knowledge cutoff of May 18, 2026. Reasoning can be switched off or set up to max, with medium as default. It supports streaming, structured outputs, function calling, prompt caching and hosted tools, but not the Realtime or fine-tuning endpoints. It is also the model in the free ChatGPT plan.

• Classifies and routes requestsLuna sorts tickets, emails or documents into categories at low cost per call. Support and operations teams can process large queues automatically.
• Extracts structured dataStructured outputs return fields such as names, dates or amounts in a fixed format. Back-office teams feed the results straight into their systems.
• Serves as a cheap step in agent chainsIn multi-step pipelines Luna handles simple steps while a larger model handles hard ones. Development teams keep running costs down.

• Priced at one hundredth of GPT-6 Astra per input token.
• Reasoning can be turned off completely for the lowest latency.

OpenAI is based in San Francisco.

Usage is billed per token, so estimate volumes and set spending limits before going live. Regional processing endpoints for data residency cost 10 percent more for models released from March 5, 2026. API data is not used for training unless you opt in, and abuse monitoring logs are kept for up to 30 days unless zero data retention is approved. In the US, check Colorado and Illinois rules if outputs feed decisions about people. For Luna, EU data residency is only available with standard processing.

No reviews yet. Be the first to share how your team uses it.

Every review shows its verification level. We read each one before it goes live. How we check reviews