GPT-6.1 Terra: benchmarks, price and rank

Its headline results

The benchmarks that count most in the rating, with its place among the models measured the same way, and where each number came from.

• APEX-Agents68.3%

  Long tasks written by investment banking analysts, management consultants and corporate lawyers from their own work, done across files and apps in a realistic work setup.

  8th of 56. Independent result from Epoch AI.

• Terminal-Bench 4.050.3%

  Hard jobs done in a real computer terminal: building and fixing software, setting up systems, and science and engineering problems.

  10th of 17. The maker's own number, from OpenAI, via LLM Stats.

• Vending-Bench 2$9,107

  The model runs a simulated vending machine business for a year on its own: finding suppliers, ordering stock, setting prices and paying the daily fees.

  13th of 60. Independent result from Epoch AI.

• SimpleBench72.2%

  Multiple-choice puzzles about everyday situations, space, time and people, written to trip up models that pattern-match instead of think. Ordinary adults do well on them.

  22nd of 68. Independent result from Epoch AI.

• SimpleQA Verified57.3%

  Short factual questions with one checked answer, asked without web search.

  22nd of 78. Independent result from Epoch AI.

• FrontierMath90.9%

  Hundreds of new, unpublished math problems written by mathematicians, from hard undergraduate level up to research, with answers a computer can check.

  5th of 76. Independent result from Epoch AI.

• LMArena WebDev1,679

  People ask two anonymous models to build the same web page or app and vote for the better one.

  12th of 98. Independent result from LMArena.

• HealthBench Professional62.7%

  Conversations in the style of real clinicians' chats with AI, graded against rubrics written by physicians.

  8th of 19. The maker's own number, from OpenAI, via LLM Stats.

• HealthBench55.1%

  5,000 realistic health conversations with patients and clinicians, each graded against criteria written by 262 physicians.

  12th of 16. The maker's own number, from OpenAI, via LLM Stats.

• LifeSci Diamond9.1%

  55 regulatory affairs tasks from drug and device makers, such as FDA and EMA submissions, labeling and compliance, with answer keys written or reviewed by practising specialists.

  3rd of 18. Independent result from Arca, via LLM Stats.