Claude Opus 5

Leaderboards

Where Claude Opus 5 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 Claude Opus 5 has a published result on, out of 100%, with its rank among the models that took it.

• Humanity's Last Exam

  #2 / 108

  Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions

  64.7%

  Self-reported · In the text score· In the vision score

• BrowseComp

  #3 / 68

  Finding hard-to-locate facts by browsing the web.

  90.8%

  Self-reported · In the agents score

• DeepSWE 1.1

  #22 / 50

  DeepSWE 1.1 evaluates software engineering agents using the mini-swe-agent harness.

  68.8%

  Self-reported · In the coding score

• Terminal-Bench 4.0

  #9 / 28

  Terminal-Bench 4.0 evaluates AI agents on real-world work in terminal and command-line containerized environments across 66 tasks, with emphasis on science-adjacent and frontier engineering problems. Relative to earlier Terminal-Bench releases, 4.0 increases timeouts and adaptively raises RAM/CPU on selected tasks to reduce harness and resource confounds.

  51.8%

  Self-reported · In the coding score· In the agents score

• AutomationBench

  #17 / 24

  AutomationBench is a tool-use benchmark that evaluates AI agents on automating real-world workflows, testing their ability to orchestrate tools and complete multi-step automation tasks.

  26.0%

  Self-reported · In the agents score

• FrontierCode 1.1

  #3 / 23

  FrontierCode 1.1 evaluates whether coding-agent changes are mergeable, using unit tests, maintainer-defined rubrics, and verifiers. Runs flagged for unfair internet use receive a zero score.

  53.4%

  Self-reported · In the coding score

• HealthBench Professional

  #12 / 19

  HealthBench Professional evaluates model capability and safety for clinician use cases using real clinician-style chats and physician-authored grading rubrics.

  59.8%

  Self-reported

• OSWorld 2.0

  #5 / 17

  OSWorld 2.0 is a benchmark of 108 long-horizon, real-world computer-use workflows spanning everyday and professional tasks. Each task is an end-to-end workflow that takes human users a median of about 1.6 hours, scored with a binary-completion metric, and targets challenges such as dynamic environments, cross-source reasoning, and implicit-state inference.

  70.6%

  Self-reported

• Legal Agent Benchmark

  #2 / 15

  The Legal Agent Benchmark (LAB) is Harvey's open-source benchmark for evaluating AI agents on complex, long-horizon legal work. Tasks are scored under an all-pass standard against expert-curated rubrics, where a task passes only if every required rubric criterion (facts, conclusions, citations, structure, and analytical moves) passes.

  11.7%

  Self-reported

• BioMysteryBench

  #1 / 7

  BioMysteryBench evaluates a model's ability to reason through challenging molecular biology problems, reporting performance on a hard subset and on the subset of problems solved by human experts.

  90.1%

  Self-reported

• ARC-AGI-3

  #2 / 5

  ARC-AGI-3 is the third-generation Abstraction and Reasoning Corpus benchmark, an interactive-reasoning evaluation designed to measure fluid, novel problem-solving ability that remains far from saturated for frontier models.

  30.2%

  Self-reported

• FrontierCode

  #2 / 4

  FrontierCode is Cognition's coding evaluation that tests whether models can pass difficult coding tasks while meeting the standards of high-quality production codebases. The Diamond subset contains the hardest problems.

  53.4%

  Self-reported

• Frontier-Bench v0.1

  #1 / 1

  Frontier-Bench v0.1 evaluates agentic terminal coding. Anthropic reports results using the mini-SWE-agent harness and a GKE backend, measured as mean reward across five attempts per task.

  43.3%

  Unverified