Use the API model ID gpt-6.1-terra to call GPT-6.1 Terra. The default reasoning effort is medium, and tool calls must use the Responses API. The model accepts text and image input, generates text, and supports the Responses tools listed on its model page.
Suitable tasks: Configuring GPT-6.1 Terra in the OpenAI API and checking its input and output modalities, reasoning effort levels, endpoint, and tool support.
Unsuitable tasks: Assessing performance, latency, or cost for a specific workload based solely on the model catalog page; the page provides no task-level test results.
Applicable model version: GPT-6.1 Terra, API model ID: gpt-6.1-terra.
Applicable client, agent, or API: OpenAI API. The model page states that tool calls require the Responses API.
Recommended reasoning effort and parameters: reasoning.effort can be set to none, low, medium, high, xhigh, or max; the default is medium. The source gives no recommended values for other request parameters.
Model configuration checklist (field names and values are compiled from the model page; this is not a complete API request example):
model = "gpt-6.1-terra"
reasoning.effort = "medium" # Default level listed on the page
tool_calling = "Use Responses API"Model capabilities listed on the page:
Input / output: Text and image input; text output.
Context window / maximum output: 1,050,000 tokens / 128,000 tokens.
Tool support listed on the model page: Web search, File search, Image generation, Code interpreter, Hosted shell, Apply patch, Skills, Computer use, MCP, and Tool search. Tool calls use the Responses API.
Features listed on the model page: Streaming, Function calling, and Structured outputs are supported; Fine-tuning is not supported.
Endpoint selection: Tool calls must use the Responses API. The site's global API navigation is not evidence that this model is compatible with other endpoints.
Select gpt-6.1-terra in the request.
Set reasoning.effort as needed for the task; if omitted, the page lists medium as the default.
If tool calls are needed, use the Responses API and select tools marked Supported on the model page.
Choose text or image input based on the input modality; the model's output is text.
The page names the model “GPT-6.1 Terra”; its code form and API ID are gpt-6.1-terra.
Reasoning effort levels are none, low, medium (default), high, xhigh, and max.
The page states: “Tool calling requires the Responses API.”
The model capability summary lists Text and Image as inputs and Text as output; the context window is 1,050,000 and the maximum output is 128,000 tokens.
All ten tools listed in the page's tool table are marked Supported. Streaming, Function calling, and Structured outputs are marked Supported; Fine-tuning is marked Not supported.
These findings record the contents of OpenAI's GPT-6.1 Terra model page as viewed on 2026-10-27; the model catalog and pricing may change.
Tool support does not mean that every tool combination, account permission, or calling scenario is available. The page explicitly requires tool calls to use the Responses API.
Other endpoints listed in the global API navigation should not be treated as part of this model's support list; check model compatibility for a specific integration.
The page provides no recommendations for other parameters such as temperature, sampling, or timeout, and no task-specific recommendations for reasoning effort; these values are not specified.
Open the original URL in the Tabbit browser and check the model name and ID, reasoning effort description, tool-calling notice, modalities and token limits, as well as the Tools, Features, and Endpoints sections. Collection date: 2026-10-27. Only content visible on that page was used.
GPT-6.1 Terra