WordOnThisWordOnThisSearch
Submit
GPT-6.1 Terra logo

GPT-6.1 Terra

Applies visual reasoning and computer use to screenshots, drawings, and UI tasks.

Visit website
GPT-6.1 Terra official website preview
Loading website screenshot…The official website preview is loading.

What is GPT-6.1 Terra?

OpenAI focuses GPT-6.1 Terra on visual reasoning and computer use alongside general software-engineering capabilities. Priced per token, it accepts text and images within a 1,050,000-token context window (up to 922,000 input and 128,000 output tokens) and produces text or structured outputs. It integrates with Responses API tools including web search, file search, and computer use. Teams can use it to interpret screenshots and technical drawings, match product images, draft text, and try screenshot-led QA or UI debugging.

What can GPT-6.1 Terra do?

Visual reasoning and image matching

Interprets screenshots and technical drawings, compares product images, and outputs text or structured data.

Computer use and search tools

Pairs with computer use, web search, and file search in the Responses API to inspect interfaces and gather context.

1,050,000-token multimodal context

Accepts text and images up to 922,000 input tokens and returns up to 128,000 output tokens per request.

Who is GPT-6.1 Terra for?

  • Screenshot-led QA and UI debuggingEngineers can try it to spot layout regressions in rendered screenshots and propose code fixes.
  • Technical drawing and catalog matchingOperations teams can match product photos against catalog records or extract details from diagrams.
  • Visual-to-text draftingWriters and support teams can draft documentation or bug reports directly from annotated captures.

How much is GPT-6.1 Terra?

View official pricing ↗
paid

OpenAI standard API, up to 272,000 input tokens: USD $1 per million uncached input tokens and $5 per million output tokens. Above 272,000 input tokens: USD $2 per million uncached input tokens and $7.50 per million output tokens for the entire request.

Frequently asked questions

What is the OpenAI API model ID?

Use gpt-6.1-terra to call this model.

How did it perform on Terminal-Bench 4.0?

Vals AI reports 49.49% on Terminal-Bench 4.0 with the Mini-SWE-agent harness. The score applies to that evaluated setup. Source: https://www.vals.ai/benchmarks/terminal-bench-4

What people are saying

Public experiences and discussions, with links to the originals.

No discussions collected yet.

Explore 11 products that share at least one second-level category.

View all 11 alternatives

Help keep this listing accurate

Found something incorrect? Suggest a correction for our editors. It’s free; no account or email is needed.