OpenAI: GPT-6 Luna (batch)

GPT-6 Luna is a text model from OpenAI, part of the GPT-6 series. Within that series it sits below GPT-6 Sol, positioned as the fast, cost-efficient option. It is aimed at high-volume and latency-sensitive work, so it is worth considering when you need many requests handled quickly rather than the absolute strongest reasoning available from the same family.

On the input side, GPT-6 Luna accepts text, images and files, and it returns text only. The context window is 1,050,000 tokens, large enough to hold very long documents or extended conversation histories in a single request. The maximum output is 128,000 tokens, so long generations are possible alongside that large input budget. It supports a range of request parameters, including reasoning controls, structured outputs and tool use, which means the model can do more than plain chat completion.

The combination of a very large context window and image and file inputs makes the model a good fit for tasks where the source material is big or mixed. You might use it to summarize or query lengthy reports and contracts, run classification or extraction over large batches of documents, or build chat experiences that need to remember a long history. Because image input is supported, it can also handle requests that combine screenshots or scanned pages with text instructions. The availability of tools and structured outputs points toward lightweight agentic workflows, where the model calls functions or returns data in a fixed schema.

You call the model through api14's OpenAI-compatible API using the model id openai/gpt-6-luna:batch. Because the interface follows the OpenAI format, existing clients and SDKs generally work with a change of base URL and model id. The supported parameters are include_reasoning, max_tokens, reasoning, reasoning_effort, response_format, seed, structured_outputs, tool_choice, tools and verbosity. These cover output length, determinism via seed, JSON or schema-constrained responses, reasoning depth and tool selection.

As for limits, the facts here do not list rate limits, quotas or a knowledge cutoff, so check the live documentation for those. The provider notes that GPT-6 Luna is positioned below GPT-6 Sol, so if a task demands the strongest reasoning in the series, a larger sibling model may fit better. For high-volume, latency-sensitive jobs, classification and lightweight agentic flows, Luna is the intended choice.