GPT-5.6 Luna

GPT-5.6 Luna is a chat-oriented large language model created by OpenAI and listed in the computeprices.com catalog as a text-generation endpoint served through API providers. It carries the GPT-5.6 version designation with the "Luna" variant name, indicating a specific configuration within OpenAI's GPT-5 series rather than a standalone architecture. The performance data we track for GPT-5.6 Luna comes from Artificial Analysis measurements: approximately 118.1 output tokens per second in sustained generation, and a time to first token of roughly 1,582 milliseconds. That combination describes a model that streams at a moderate-to-fast rate once generation begins, but has a noticeable initial latency of over a second and a half before the first token arrives — a profile common to larger models and to deployments that perform additional pre-generation work. For interactive chat, the sub-two-second first-token figure is generally acceptable; for latency-critical pipelines that fire many short requests, the fixed startup cost is the more relevant number than the streaming rate. Beyond the creator, model type, and these throughput measurements, we do not currently track confirmed details for GPT-5.6 Luna such as context window size, input modalities, tool-calling support, or published benchmark scores. Readers evaluating it against other OpenAI models or third-party alternatives should check the serving provider's own documentation for those specifications, and use the pricing table on this page to compare what individual providers charge to serve it.