OpenAI: GPT-5.6 Terra

GPT-5.6 Terra is a text model from OpenAI, part of the GPT-5.6 series. OpenAI positions it as the balanced tier of that series, sitting between the flagship Sol tier and the cost-efficient Luna tier. It was released by the provider on 9 July 2026, and its knowledge cutoff is 16 February 2026. If you are choosing within the GPT-5.6 family, Terra is the middle option: more capable than Luna, less capable than Sol, and intended for general work rather than a single narrow niche.

The model accepts text, images and files as input, and it returns text only. Its context window is 1,050,000 tokens, which is large enough to hold entire codebases, long document sets or extended conversation histories in a single request. The maximum output is 128,000 tokens, so responses can be long as well as the inputs. Because image and file inputs are supported alongside text, you can mix formats in one prompt rather than converting everything to plain text first.

Those capabilities point to a few practical uses. Everyday coding assistance fits well: reading a repository or a set of source files as input and producing patches, explanations or new code as output. Reasoning tasks over long material also fit, since the context window allows you to supply full documents instead of excerpts. Agentic workflows are another match, given the tool calling parameters listed below. Document and image review is possible too, for example extracting structured data from scanned pages or screenshots and returning it as text.

You call the model through api14's OpenAI-compatible API using the model id "openai/gpt-5.6-terra". The supported request parameters are include_reasoning, max_completion_tokens, max_tokens, reasoning, reasoning_effort, response_format, seed, structured_outputs, tool_choice, tools and verbosity. In practice that means you can control output length, ask for reasoning traces or set how much reasoning effort is spent, request JSON or other structured formats, set a seed for reproducibility, and define tools with a tool choice policy for agent-style loops. The verbosity parameter lets you steer how long or short the prose output is.

One boundary worth noting is that output is text only, so if you need generated images or audio, this model is not the right fit. The provider description frames Terra around everyday coding, reasoning and agentic work, which suggests that for the most demanding tasks in those areas the flagship Sol tier may be the better choice, while Luna may suit lighter workloads. Within those bounds, Terra covers a wide range of text-in, text-out applications.