GPT-4

Leaderboards

Where GPT-4 sits among the top 20 in each category.

By task

Scored from votes on each kind of prompt alone, on the same 1,000-point scale.

Test results

Every test GPT-4 has a published result on, out of 100%, with its rank among the models that took it.

• GPQA

  #236 / 256

  Graduate-level science questions written to be hard to look up.

  35.7%

  Self-reported

• MMLU

  #31 / 103

  Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains

  86.4%

  Self-reported

• MATH

  #66 / 71

  MATH dataset contains 12,500 challenging competition mathematics problems from AMC 10, AMC 12, AIME, and other mathematics competitions. Each problem includes full step-by-step solutions and spans multiple difficulty levels (1-5) across seven mathematical subjects including Prealgebra, Algebra, Number Theory, Counting and Probability, Geometry, Intermediate Algebra, and Precalculus.

  42.0%

  Self-reported

• HumanEval

  #59 / 66

  A benchmark that measures functional correctness for synthesizing programs from docstrings, consisting of 164 original programming problems assessing language comprehension, algorithms, and simple mathematics

  67.0%

  Self-reported

• MGSM

  #21 / 31

  MGSM (Multilingual Grade School Math) is a benchmark of grade-school math problems. Contains 250 grade-school math problems manually translated from the GSM8K dataset into ten typologically diverse languages: Spanish, French, German, Russian, Chinese, Japanese, Thai, Swahili, Bengali, and Telugu. Evaluates multilingual mathematical reasoning capabilities.

  74.5%

  Self-reported

• DROP

  #11 / 30

  DROP (Discrete Reasoning Over Paragraphs) is a reading comprehension benchmark requiring discrete reasoning over paragraph content. It contains crowdsourced, adversarially-created questions that require resolving references and performing discrete operations like addition, counting, or sorting, demanding comprehensive paragraph understanding beyond paraphrase-and-entity-typing shortcuts.

  80.9%

  Self-reported

• HellaSwag

  #2 / 28

  A challenging commonsense natural language inference dataset that uses Adversarial Filtering to create questions trivial for humans (>95% accuracy) but difficult for state-of-the-art models, requiring completion of sentence endings based on physical situations and everyday activities

  95.3%

  Self-reported

• Winogrande

  #1 / 23

  WinoGrande: An Adversarial Winograd Schema Challenge at Scale. A large-scale dataset of 44,000 pronoun resolution problems designed to test machine commonsense reasoning. Uses adversarial filtering to reduce spurious biases and provides a more robust evaluation of whether AI systems truly understand commonsense or exploit statistical shortcuts. Current best AI methods achieve 59.4-79.1% accuracy, significantly below human performance of 94.0%.

  87.5%

  Self-reported

• AI2 Reasoning Challenge (ARC)

  #1 / 1

  A dataset of 7,787 genuine grade-school level, multiple-choice science questions assembled to encourage research in advanced question-answering. The dataset is partitioned into a Challenge Set and Easy Set, where the Challenge Set contains only questions answered incorrectly by both retrieval-based and word co-occurrence algorithms. Covers multiple scientific domains including biology, physics, earth science, and chemistry, requiring scientific reasoning, causal understanding, and conceptual knowledge beyond simple fact retrieval. Includes a supporting corpus of over 14 million science sentences.

  96.3%

  Unverified

• LSAT

  #1 / 1

  LSAT (Law School Admission Test) benchmark evaluating complex reasoning capabilities across three challenging tasks: analytical reasoning, logical reasoning, and reading comprehension. The LSAT measures skills considered essential for success in law school including critical thinking, reading comprehension of complex texts, and analysis of arguments.

  88.0%

  Unverified

• SAT Math

  #1 / 1

  SAT Math benchmark from AGIEval containing standardized mathematics questions from the College Board SAT examination, designed to evaluate mathematical reasoning capabilities of foundation models using human-centric assessment methods.

  89.0%

  Unverified

• Uniform Bar Exam

  #1 / 1

  The Uniform Bar Examination (UBE) benchmark evaluates language models on the complete bar exam including multiple-choice Multistate Bar Examination (MBE), open-ended Multistate Essay Exam (MEE), and Multistate Performance Test (MPT) components. Used to assess legal reasoning capabilities across seven subject areas including Evidence, Torts, Constitutional Law, Contracts, Criminal Law and Procedure, Real Property, and Civil Procedure.

  90.0%

  Unverified