OpenAI has released GPT-6.1 Terra, a fast model built for visual reasoning, computer use and writing, priced at $1 per million input tokens and $5 per million output tokens — half the price of GPT-6.1 Sol. The new model replaces GPT-5.6 Terra and is live in the API as gpt-6.1-terra, aimed squarely at teams automating screens, forms and documents at volume.
GPT-6.1 Terra (Credits: MarkTechPost)
GPT-6.1 Terra price and performance: $0.29 per agent task
The headline number is cost per completed job. On OpenAI’s Agents’ Last Exam v1, a set of professional computer tasks that includes turning architectural drawings into a 3D model, Terra scores 57.6% at medium effort for $0.29 per task. GPT-6 Astra matches that score but costs $4.82 per task, while the previous GPT-5.6 Terra managed 42.6% at $0.84. At max effort, Terra reaches 60.0% at $1.01 per task, and OpenAI says it completes tasks more than three times faster than GPT-6.1 Sol and GPT-6 Astra.
Terra accepts text and image input and returns text, with a 1,050,000-token context window and up to 128,000 output tokens. Its knowledge cutoff is June 1, 2026. Tool calling requires the Responses API, with supported tools including computer use, hosted shell, code interpreter, web search, file search and MCP. OpenAI lists use cases such as matching product photos to catalog entries and interpreting technical drawings.
GPT-6.1 Terra benchmarks: where it wins and where it trails
On MMMU-Pro (Vision), Terra scores 85.5% at max effort, just behind GPT-6.1 Sol at 86.3% and GPT-6 Astra at 86.7%, but at $0.012 per task against Astra’s $0.065. On the OSWorld 2.1 offline set it scores 66.7% at medium and 74.2% at max, up from GPT-5.6 Terra’s peak of 57.1%, though still behind Sol (78.0%), Astra (80.9%) and Claude Sonnet 5.5 (83.9%).
There are limits. On Terminal-Bench 4.0, Terra solves 50.3% at max effort — a big jump from GPT-5.6 Terra’s 21.5%, but behind Sol (55.9%), Astra (58.2%) and Claude Sonnet 5.5 (70.6%). OpenAI itself recommends Sol or Astra for the most complex coding tasks. On the brighter side, error rates on OpenAI’s internal set of error-inducing prompts fell from 53.6% to 16.7% at low effort, and to 4.3% at max effort.
Standard API pricing is $1 input, $0.05 cached input and $5 output per million tokens, with Batch and Flex 50% cheaper and Fast mode at twice the standard rate. In ChatGPT, Terra is live for Plus, Pro, Enterprise and Edu users in ChatGPT Work and Codex. For agent workloads where volume matters more than squeezing out the last few benchmark points, Terra’s pitch is simple: near-flagship results at a fraction of the bill.
Also Read: How OpenAI Turned Its Coding Tools Into Dots, an Always-On Personal Assistant | Tsinghua Open-Sources AKV, a Cache Engineering Server That Cuts LLM Agent Costs 27.7%
Source: MarkTechPost — read the full story there.
