RLTasksby ulam.ai
Models / OpenAI

OpenAI / MODEL EVIDENCE

GPT-5 nano

35 benchmarks · 62 observations · 38 sources · released 2025-08-07

Scores at a glance

35 benchmarks · large dot: GPT-5 nano; small dots: closest models where reported

MATH Level 5 ↗95.24%
95.2495.24

GPT-5 nano (medium) · MATH level 5 · Best score (across scorers) · % ↗

1 more reported result

94.9% · GPT-5 nano (high) · MATH level 5 · Best score (across scorers) · % · Epoch AI · MATH level 5 ↗

Terminal-Bench 2.0 ↗21.8%
3.0784.72

GPT-5 nano (unknown thinking) · Terminal Bench · Accuracy mean · % ↗

8 more reported results

11.46% · GPT-5 nano (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

9.89% · GPT-5 nano (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

7.87% · GPT-5 nano (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

6.97% · GPT-5 nano (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

11.5% · GPT-5 nano (medium) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

9.9% · GPT-5 nano (medium) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

7.9% · GPT-5 nano (medium) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

7% · GPT-5 nano (medium) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

VPCT ↗37.2%
37.237.2

GPT-5 nano (high) · VPCT · Correct · % ↗

1 more reported result

35.4% · GPT-5 nano (medium) · VPCT · Correct · % · Epoch AI · VPCT ↗

LMCA ↗7.93%
2.7868.26

GPT-5 nano (high) · LMCA · Score · % ↗

DTBench ↗62.67%
41.6598.93

GPT-5 nano (high) · DTBench · Accuracy · % ↗

EconBench ↗0.0636 error points
0.060.06

gpt-5-nano · EconBench · EconBench prototype · risk-consistency error · published prototype observations ↗

Closest models by score

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

16.9 percentile-point gap

11 of 16 shared benchmark comparisons

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. CMT-Benchmark: A Benchmark for Condensed Matter Theory Built by Expert Researchers — published result table ↗

    Pan, Haining, Roggeveen, James V., Berg, Erez, Carrasquilla, Juan, Chowdhury, Debanjan, Ganguli, Surya et al. · 2025

    All 19 authors

    Pan, Haining, Roggeveen, James V., Berg, Erez, Carrasquilla, Juan, Chowdhury, Debanjan, Ganguli, Surya, Ghimenti, Federico, Hasik, Juraj, Hunt, Henry, Jiang, Hong-Chen, Kamb, Mason, Kao, Ying-Jer, Khatami, Ehsan, Lawler, Michael J., Luo, Di, Neupert, Titus, Qi, Xiaoliang, Brenner, Michael P., Kim, Eun-Ah

  2. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  3. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  4. EconBench: Economic choice by language models — gpt-5-nano prototype metrics ↗

    EconBench project · 2026

  5. ARC-AGI-1 ↗

    Epoch AI

  6. ARC-AGI-2 ↗

    Epoch AI

  7. Chess Puzzles ↗

    Epoch AI

  8. FrontierMath Tier 4 (v1, old) ↗

    Epoch AI

  9. FrontierMath Tier 4 (v2) ↗

    Epoch AI

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

    Epoch AI

  11. FrontierMath Tiers 1-3 (v2) ↗

    Epoch AI

  12. GPQA Diamond ↗

    Epoch AI

  13. MATH Level 5 ↗

    Epoch AI

  14. Mystery Game Puzzles ↗

    Epoch AI

  15. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  16. SimpleQA Verified ↗

    Epoch AI

  17. Terminal-Bench 2.0 ↗

    Epoch AI

  18. Capabilities & benchmarking — retained external evaluation records ↗

    Epoch AI · 2026

  19. AI benchmarks & capabilities ↗

    Epoch AI

  20. Fiction.liveBench — March 14, 2025 benchmark release ↗

    Fiction.live · 2025

  21. WeirdML ↗

    Håvard Tveit Ihle

  22. Terminal-Bench 2.0 leaderboard ↗

    Terminal-Bench Team

  23. AIME — evaluation methodology and results ↗

    Vals AI · 2026

  24. CaseLaw — evaluation methodology and results ↗

    Vals AI · 2026

  25. GPQA Diamond — evaluation methodology and results ↗

    Vals AI · 2026

  26. LiveCodeBench — evaluation methodology and results ↗

    Vals AI · 2026

  27. LegalBench — evaluation methodology and results ↗

    Vals AI · 2026

  28. MATH 500 — evaluation methodology and results ↗

    Vals AI · 2026

  29. MedCode — evaluation methodology and results ↗

    Vals AI · 2026

  30. MedQA — evaluation methodology and results ↗

    Vals AI · 2026

  31. MedScribe — evaluation methodology and results ↗

    Vals AI · 2026

  32. MGSM — evaluation methodology and results ↗

    Vals AI · 2026

  33. MMLU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  34. MMMU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  35. MortgageTax — evaluation methodology and results ↗

    Vals AI · 2026

  36. ProofBench — evaluation methodology and results ↗

    Vals AI · 2026

  37. SAGE — evaluation methodology and results ↗

    Vals AI · 2026

  38. TaxEval — evaluation methodology and results ↗

    Vals AI · 2026

Names used in source tables

GPT 5 Nano · GPT-5 nano · GPT-5 nano (high) · GPT-5 nano (low) · GPT-5 nano (medium) · GPT-5 nano (unknown thinking) · GPT-5-nano · gpt-5-nano