GPT-5.6 Luna
GPT-5.6 Luna is OpenAI's cost-optimized GPT-5.6 model for high-volume workloads, with a 1.05-million-token context window and up to 128K output tokens.
Overview
The model accepts text and image input and returns text. It supports reasoning controls, function calling, structured output, code execution, computer use, and other OpenAI tool surfaces.
Model specifications
- Family
- GPT-5.6
- Context
- 1,050,000 tokens
- Max output
- 128,000 tokens
- Input
- Text, Image
- Output
- Text
- Open weights
- No
- Capabilities
- Reasoning, Vision, Tool Use, Function Calling, Structured Output, Code Execution, Computer Use
Recorded API token pricing
Reference rates in USD per 1 million tokens. Actual charges depend on the provider, service tier, context length, tools, and region.
- Input
- USD 0.2 / 1M tokens
- Cached input
- USD 0.02 / 1M tokens
- Output
- USD 1.2 / 1M tokens
Pricing caveats
Current standard text-token pricing shown on the OpenAI model page. Prompts above 272K input tokens use the documented long-context pricing multiplier.
Reviewed Oct 4, 2026 · Verify with provider documentation →
Related entities
Devin CLI
Devin CLI is Cognition's local coding agent for terminal workflows. It can work with multiple model providers, run against a local repository, and hand sessions off to Devin Cloud.
Droid
Droid is Factory's software-development agent for planning, implementing, reviewing, and validating code changes across desktop, terminal, web, and managed execution environments.
GitHub Copilot
GitHub Copilot provides code completion, chat, agent mode, code review, a cloud coding agent, and a command-line interface across GitHub and supported development environments.