GPT-6 Luna
GPT-6 Luna is intended for focused tasks at high volume. Its tool-calling behavior depends on both the endpoint and the reasoning setting, a detail that matters when replacing another model.
The low base rate has a context condition
Prompts above 272,000 input tokens double input and cache rates and multiply output rates by 1.5 for the full request. EU data residency is offered for Standard, Fast, Flex and Batch, subject to eligibility; regional processing adds a premium.GPT-6 Luna model reference
Give a small model a well-defined decision
Practical guidance. For incoming support messages, define a short list of categories and include an uncertain outcome. Measure missed urgent cases separately from overall accuracy. A model can be useful for routing even when complex case resolution belongs with a different workflow.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API |
| Model identifiers | gpt-6-luna |
| Context window | 1,050,000 tokens API context limit GPT-6 Luna model reference |
| Maximum output | 128,000 tokens Maximum output GPT-6 Luna model reference |
| Input price | 0.1 USD per million tokens Standard text input; up to 272,000 input tokens GPT-6 Luna model reference |
| Output price | 0.5 USD per million tokens Standard text output; up to 272,000 input tokens GPT-6 Luna model reference |
| Input and output | Text and images in; text outGPT-6 Luna model reference |
| Chat Completions tool calls | Function calling requires reasoning_effort set to none.GPT-6 Luna model reference |
| Responses API | Supports built-in tools and function calling.GPT-6 Luna model reference |
| Maximum input | 922,000 tokens within the total context window.GPT-6 Luna model reference |
Related models
- GPT-6.1 Sol
OpenAI · GPT-6.1
A model for coding, tool use and complex professional tasks.
Launch date not confirmedAPI - GPT-6.1 Terra
OpenAI · GPT-6.1
A model for visual reasoning, screen interaction and writing with optional reasoning.
Launch date not confirmedAPI