GPT-5.6 Luna

GPT-5.6 Luna is a OpenAI model available in Agent Serve. GPT-5.6 Luna is a OpenAI model tracked in Agent Serve. It supports a 1.05M token context window. Pricing starts at $0.2/1M input tokens and $1.2/1M output tokens. Key capabilities include Tool choice, Structured outputs, Reasoning none, low, medium, high, xhigh, max.

GPT-5.6 Luna starts at $0.2/1M input tokens, $0.02/1M cached input tokens, and $1.2/1M output tokens. Above 272k input tokens, the full request is priced at $0.4/1M input tokens, $0.04/1M cached input tokens, and $1.8/1M output tokens.

GPT-5.6 Luna supports a context window of 1.05M tokens in Agent Serve. In an agent, this determines how much conversation history, tool outputs, and retrieved documents the model can hold in a single call.

GPT-5.6 Luna supports the following capabilities in Agent Serve: Tool choice, Structured outputs, Reasoning none, low, medium, high, xhigh, max, Verbosity low, medium, high, Max output 128k.

Best for reasoning-heavy tasks that need more deliberate model control. When used in an Agent Serve workflow, it can be selected in any Agent block from the model picker.