RLTasksby ulam.ai
Models / Anthropic

Anthropic / MODEL EVIDENCE

Claude Opus 4.1

38 benchmarks · 74 observations · 44 sources · released 2025-08-05

Scores at a glance

38 benchmarks · large dot: Claude Opus 4.1; small dots: closest models where reported

GPQA Diamond ↗77.27%
77.2777.27

Claude Opus 4.1 · GPQA diamond · Best score (across scorers) · % ↗

4 more reported results

76.77% · Claude Opus 4.1 · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗

73.23% · Claude Opus 4.1 · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗

76.26 % · Claude Opus 4.1 (Thinking) · GPQA Diamond · Accuracy · Vals AI · GPQA Diamond ↗

69.95 % · Claude Opus 4.1 (Nonthinking) · GPQA Diamond · Accuracy · Vals AI · GPQA Diamond ↗

Cybench ↗42%
555

Claude Opus 4.1 · Cybench · Unguided % Solved · % ↗

Terminal-Bench 2.0 ↗38%
3.178.4

Claude Opus 4.1 · Terminal Bench · Accuracy mean · % ↗

7 more reported results

36.9% · Claude Opus 4.1 · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

35.1% · Claude Opus 4.1 · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

34.8% · Claude Opus 4.1 · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

37.98% · Claude Opus 4.1 (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

36.85% · Claude Opus 4.1 (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

35.06% · Claude Opus 4.1 (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

34.83% · Claude Opus 4.1 (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

GDPval ↗43.6%
9.949.7

Claude Opus 4.1 · GDPval · Win Rate (%) · % ↗

METR task performance & time horizons ↗66.81%
10.185.2

Claude Opus 4.1 · METR task performance & time horizons · Average task accuracy ↗

3 more reported results

61.57% · Claude Opus 4.1 · METR task performance & time horizons · Average task accuracy · Epoch AI · METR Time Horizons ↗

100.472 minutes · Claude Opus 4.1 · METR task performance & time horizons · 50% success task horizon · METR-Horizon-v1.1 · Epoch AI · METR Autonomy / Task-Horizon Suite · public result data ↗

113.68507 minutes · Claude Opus 4.1 · METR task performance & time horizons · 50% success task horizon · Epoch AI · METR Autonomy / Task-Horizon Suite · public result data ↗

LMCA ↗37.08%
2.7868.26

Claude Opus 4.1 (unknown thinking) · LMCA · Score · % ↗

DTBench ↗80%
41.6598.93

Claude Opus 4.1 (unknown thinking) · DTBench · Accuracy · % ↗

EnigmaEval ↗7.18%
0.3839.28

Claude Opus 4.1 (unknown thinking) · EnigmaEval · Accuracy · % ↗

LegalBench ↗83.46 %
32.9788.56

Claude Opus 4.1 (Nonthinking) · LegalBench · Accuracy ↗

MedCode ↗47.23 %
47.2347.23

Claude Opus 4.1 (Thinking) · MedCode · Accuracy ↗

1 more reported result

41.37 % · Claude Opus 4.1 (Nonthinking) · MedCode · Accuracy · Vals AI · MedCode ↗

MATH-500 ↗95.4 %
95.495.4

Claude Opus 4.1 (Thinking) · MATH 500 · Accuracy ↗

1 more reported result

93 % · Claude Opus 4.1 (Nonthinking) · MATH 500 · Accuracy · Vals AI · MATH 500 ↗

MMLU-Pro ↗87.92 %
87.9287.92

Claude Opus 4.1 (Thinking) · MMLU Pro · Accuracy ↗

1 more reported result

87.21 % · Claude Opus 4.1 (Nonthinking) · MMLU Pro · Accuracy · Vals AI · MMLU Pro ↗

MMMU-Pro ↗77.51 %
77.5177.51

Claude Opus 4.1 (Thinking) · MMMU Pro · Accuracy ↗

1 more reported result

73.72 % · Claude Opus 4.1 (Nonthinking) · MMMU Pro · Accuracy · Vals AI · MMMU Pro ↗

AlgoTune ↗1.34 ×
1.312.05

Claude Opus 4.1 (high) · Harmonic-mean speedup ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

How similarity is measured

Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. Arena leaderboard — WebDev evaluation ↗

    Arena

  2. TextQuests: How Good are LLMs at Text-Based Video Games? — published result table ↗

    Phan, Long, Mazeika, Mantas, Zou, Andy, Hendrycks, Dan · 2025

  3. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  4. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  5. Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models ↗

    Zhang, Andy K., Perry, Neil, Dulepet, Riya, Ji, Joey, Menders, Celeste, Lin, Justin W. et al. · 2024

    All 27 authors

    Zhang, Andy K., Perry, Neil, Dulepet, Riya, Ji, Joey, Menders, Celeste, Lin, Justin W., Jones, Eliot, Hussein, Gashon, Liu, Samantha, Jasper, Donovan, Peetathawatchai, Pura, Glenn, Ari, Sivashankar, Vikram, Zamoshchin, Daniel, Glikbarg, Leo, Askaryar, Derek, Yang, Mike, Zhang, Teddy, Alluri, Rishi, Tran, Nathan, Sangpisit, Rinnara, Yiorkadjis, Polycarpos, Osele, Kenny, Raghupathi, Gautham, Boneh, Dan, Ho, Daniel E., Liang, Percy

  6. FutureSearch Evals ↗

    FutureSearch

  7. Algotune — methodology and published results ↗

    Epoch AI · 2026

  8. Earthborne Rangers (EBR-bench) ↗

    Epoch AI

  9. FrontierMath Tier 4 (v1, old) ↗

    Epoch AI

  10. FrontierMath Tier 4 (v2) ↗

    Epoch AI

  11. FrontierMath Tiers 1-3 (v1, old) ↗

    Epoch AI

  12. FrontierMath Tiers 1-3 (v2) ↗

    Epoch AI

  13. GDPval ↗

    Epoch AI

  14. GPQA Diamond ↗

    Epoch AI

  15. Humanity's Last Exam ↗

    Epoch AI

  16. METR Time Horizons ↗

    Epoch AI

  17. Mystery Game Puzzles ↗

    Epoch AI

  18. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  19. SimpleBench ↗

    Epoch AI

  20. SWE-bench Verified ↗

    Epoch AI

  21. Terminal-Bench 2.0 ↗

    Epoch AI

  22. Capabilities & benchmarking — retained external evaluation records ↗

    Epoch AI · 2026

  23. Text Arena (Coding) ↗

    Epoch AI

  24. AI benchmarks & capabilities ↗

    Epoch AI

  25. Writing benchmark — incorporating mandatory story elements ↗

    lechmazur

  26. WeirdML ↗

    Håvard Tveit Ihle

  27. EnigmaEval ↗

    Scale Labs / SEAL

  28. Measuring AI Ability to Complete Long Software Tasks ↗

    METR

  29. Metr Autonomy Task Horizon Suite — methodology and published results ↗

    METR · 2026

  30. Terminal-Bench 2.0 leaderboard ↗

    Terminal-Bench Team

  31. AIME — evaluation methodology and results ↗

    Vals AI · 2026

  32. GPQA Diamond — evaluation methodology and results ↗

    Vals AI · 2026

  33. LiveCodeBench — evaluation methodology and results ↗

    Vals AI · 2026

  34. LegalBench — evaluation methodology and results ↗

    Vals AI · 2026

  35. MATH 500 — evaluation methodology and results ↗

    Vals AI · 2026

  36. MedCode — evaluation methodology and results ↗

    Vals AI · 2026

  37. MedQA — evaluation methodology and results ↗

    Vals AI · 2026

  38. MedScribe — evaluation methodology and results ↗

    Vals AI · 2026

  39. MGSM — evaluation methodology and results ↗

    Vals AI · 2026

  40. MMLU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  41. MMMU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  42. MortgageTax — evaluation methodology and results ↗

    Vals AI · 2026

  43. SAGE — evaluation methodology and results ↗

    Vals AI · 2026

  44. TaxEval — evaluation methodology and results ↗

    Vals AI · 2026

Names used in source tables

Claude Opus 4.1 · Claude Opus 4.1 (Nonthinking) · Claude Opus 4.1 (Thinking) · Claude Opus 4.1 (high) · Claude Opus 4.1 (unknown thinking)