GPT-6.1 Terra

Not yet independently verified. Rests on OpenAI's own API changelog and model page; OpenAI's announcement page could not be read (403) and no independent dated test was found. Capability claims are OpenAI's own. We will update this when it can be confirmed, and remove this note.

GPT-6.1 Terra is a large language model that OpenAI released on 26 October 2026 as gpt-6.1-terra on its Responses and Chat Completions APIs, according to the API changelog of that day. OpenAI describes it as a fast, efficient model for visual reasoning, precise computer use and clear, natural writing. It sits below GPT-6.1 Sol in OpenAI's GPT-6 line and costs half as much per token.

What it does

The model takes text and images as input and returns text. OpenAI's model page suggests uses such as matching product photos to catalogue entries, interpreting technical drawings, acting on application screens and drafting or revising text. Reasoning effort can be set from none through low, medium (the default), high and xhigh to max, and the model supports streaming, function calling and structured outputs. Tool calling requires the Responses API, where web search, file search, image generation and the code interpreter are available as tools. Fine-tuning is not supported.

Pricing and limits

The context window is 1,050,000 tokens with up to 128,000 output tokens, and the knowledge cutoff is 1 June 2026. Standard pricing is $1.00 per million input tokens, $0.05 for cached input and $5.00 for output. Cache writes cost $1.25 per million tokens for 30-minute retention and $2.50 for 12 hours. Prompts over 272,000 input tokens are charged at twice the input and cache rates and 1.5 times the output rate for the whole request. Batch and Flex prices are half the standard rate, regional processing adds 10% where offered, and fast mode costs twice the standard rate.

Who it is for

Terra suits teams building agents that read screens or documents with images, and high-volume writing or extraction jobs where GPT-6.1 Sol's price is hard to justify. Teams already on GPT-6 models can move between them with the same APIs; OpenAI points to its GPT-6 guide for prompting and migration.

Key facts

Maker: OpenAI Released: 26 October 2026 Context: 1,050,000 tokens, 128,000 output Knowledge cutoff: 1 June 2026 Price: $1.00 input, $0.05 cached input, $5.00 output per 1M tokens Input: text and image; output text Licence: proprietary, API only