OpenAI: GPT-5.6 Luna (batch)

GPT-5.6 Luna (batch) is an OpenAI model that accepts text, image, and file inputs. It offers a context window of 1,050,000 tokens and can return up to 128,000 tokens in a single completion, which suits long documents or large multi-file inputs. It supports tool calling and reasoning. The listing does not specify whether it supports structured output or whether its weights are open. At $0.10 per million input tokens and $0.60 per million output tokens, it sits at the low end of paid pricing. That makes it worth shortlisting for high-volume or long-context workloads where cost per token matters. Its blended benchmark score is 89.4, with a coding index of 100.0, an overall AI index of 84.5, and an agentic index of 75.2. Coding is its clearest strength. However, that blend covers only three benchmarks, so teams should validate it on their own tasks, especially agentic ones, before committing.