RLTasksby ulam.ai
Models / OpenAI

OpenAI / MODEL EVIDENCE

GPT-5

66 benchmarks · 136 observations · 76 sources · released 2025-08-07

Scores at a glance

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

Aider coding benchmarks ↗88%
8888

GPT-5 (high) · Aider coding benchmarks · Polyglot · percent correct ↗

5 more reported results

86.7% · GPT-5 (medium) · Aider coding benchmarks · Polyglot · percent correct · Epoch AI · Aider polyglot ↗

81.3% · GPT-5 (low) · Aider coding benchmarks · Polyglot · percent correct · Epoch AI · Aider polyglot ↗

88% · gpt-5 (high) · Aider coding benchmarks · Polyglot · percent correct · Aider code editing benchmark ↗

86.7% · gpt-5 (medium) · Aider coding benchmarks · Polyglot · percent correct · Aider code editing benchmark ↗

81.3% · gpt-5 (low) · Aider coding benchmarks · Polyglot · percent correct · Aider code editing benchmark ↗

GSO ↗6.86%
026.5

GPT-5 (high) · GSO-Bench · Score OPT@1 · % ↗

Terminal-Bench 2.0 ↗49.61%
3.0784.72

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

7 more reported results

43.82% · GPT-5 (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

35.17% · GPT-5 (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

33.93% · GPT-5 (unknown thinking) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

49.6% · GPT-5 (medium) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

43.8% · GPT-5 (medium) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

35.2% · GPT-5 (medium) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

33.9% · GPT-5 (medium) · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗

GDPval ↗34.8%
9.949.7

GPT-5 (medium) · GDPval · Win Rate (%) · % ↗

METR task performance & time horizons ↗69.61%
10.185.2

GPT-5 (medium) · METR task performance & time horizons · Average task accuracy ↗

3 more reported results

69.37% · GPT-5 (high) · METR task performance & time horizons · Average task accuracy · Epoch AI · METR Time Horizons ↗

203.01258 minutes · GPT-5 (high) · METR task performance & time horizons · 50% success task horizon · METR-Horizon-v1.1 · Epoch AI · METR Autonomy / Task-Horizon Suite · public result data ↗

137.31854 minutes · GPT-5 (medium) · METR task performance & time horizons · 50% success task horizon · Epoch AI · METR Autonomy / Task-Horizon Suite · public result data ↗

LMCA ↗39.99%
2.7868.26

GPT-5 (high) · LMCA · Score · % ↗

DTBench ↗90.67%
41.6598.93

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

SciCode ↗42.94%
1.2766.9

GPT-5 (unknown thinking) · SciCode · Score · % ↗

EnigmaEval ↗10.47%
0.3839.28

GPT-5 (unknown thinking) · EnigmaEval · Accuracy · % ↗

GAIA-2 ↗52.7%
7.169.2

GPT-5 (low) · GAIA-2 · Gaia2 full · Execution pass@1 · baseline ReAct · three runs ↗

15 more reported results

69.2% · GPT-5 (high) · GAIA-2 · Gaia2 full · Execution pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

64.2% · GPT-5 (low) · GAIA-2 · Gaia2 full · Search pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

79.6% · GPT-5 (high) · GAIA-2 · Gaia2 full · Search pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

39.6% · GPT-5 (low) · GAIA-2 · Gaia2 full · Ambiguity pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

51.9% · GPT-5 (high) · GAIA-2 · Gaia2 full · Ambiguity pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

30.2% · GPT-5 (low) · GAIA-2 · Gaia2 full · Adaptability pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

40.4% · GPT-5 (high) · GAIA-2 · Gaia2 full · Adaptability pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

2.3% · GPT-5 (low) · GAIA-2 · Gaia2 full · Time pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

0% · GPT-5 (high) · GAIA-2 · Gaia2 full · Time pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

28.3% · GPT-5 (low) · GAIA-2 · Gaia2 full · Noise pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

35.4% · GPT-5 (high) · GAIA-2 · Gaia2 full · Noise pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

24.6% · GPT-5 (low) · GAIA-2 · Gaia2 full · Agent2Agent pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

17.9% · GPT-5 (high) · GAIA-2 · Gaia2 full · Agent2Agent pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

34.6% · GPT-5 (low) · GAIA-2 · Gaia2 full · Overall pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

42.1% · GPT-5 (high) · GAIA-2 · Gaia2 full · Overall pass@1 · baseline ReAct · three runs · Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments ↗

Biomni-Eval ↗0.546 / 1
0.320.67

GPT-5 · Biomni-Eval · Biomni-Eval1 · mean task reward · Biomni-R0 · A1 scaffold and E1 environment ↗

10 more reported results

0 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · CRISPR delivery · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

0.62 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · GWAS causal gene · catalog · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

0.74 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · GWAS causal gene · OpenTargets · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

0.62 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · GWAS causal gene · PharmaProjects · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

0.63 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · GWAS variant prioritization · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

0.54 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · LAB-Bench DbQA · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

0.72 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · LAB-Bench SeqQA · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

0.4 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · Patient gene detection · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

0.63 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · Rare disease diagnosis · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

0.56 / 1 · GPT-5 · Biomni-Eval · Biomni-Eval1 · Screen gene retrieval · task reward · Biomni-R0 · A1 scaffold and E1 environment · Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

EconBench ↗0.0684 error points
0.070.07

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

ALE-Bench ↗807.65 Performance points
137.782951.32

gpt-5 · Average contest performance ↗

Closest models by score

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

9.2 percentile-point gap

18 of 25 shared benchmark comparisons

o3 →OpenAI
12.1 percentile-point gap

17 of 25 shared benchmark comparisons

12.4 percentile-point gap

15 of 25 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. Aider code editing benchmark ↗

    Paul Gauthier, Aider contributors

  2. Aider LLM Leaderboards — Polyglot coding benchmark ↗

    Aider

  3. ARC Prize - Leaderboard ↗

    ARC Prize Foundation

  4. Arena leaderboard — WebDev evaluation ↗

    Arena

  5. AIME 2025 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  6. CritPt v2 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  7. Global-MMLU-Lite Benchmark | Artificial Analysis ↗

    Artificial Analysis

  8. LiveCodeBench Benchmark | Artificial Analysis ↗

    Artificial Analysis

  9. MATH-500 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  10. MMLU-Pro Benchmark | Artificial Analysis ↗

    Artificial Analysis

  11. SciCode Benchmark | Artificial Analysis ↗

    Artificial Analysis

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

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

  13. 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

  14. DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models — published result table ↗

    DeepSeek-AI, Liu, Aixin, Mei, Aoxue, Lin, Bangcai, Xue, Bing, Wang, Bingxuan et al. · 2025

    All 264 authors

    DeepSeek-AI, Liu, Aixin, Mei, Aoxue, Lin, Bangcai, Xue, Bing, Wang, Bingxuan, Xu, Bingzheng, Wu, Bochao, Zhang, Bowei, Lin, Chaofan, Dong, Chen, Lu, Chengda, Zhao, Chenggang, Deng, Chengqi, Xu, Chenhao, Ruan, Chong, Dai, Damai, Guo, Daya, Yang, Dejian, Chen, Deli, Li, Erhang, Zhou, Fangqi, Lin, Fangyun, Dai, Fucong, Hao, Guangbo, Chen, Guanting, Li, Guowei, Zhang, H., Xu, Hanwei, Li, Hao, Liang, Haofen, Wei, Haoran, Zhang, Haowei, Luo, Haowen, Ji, Haozhe, Ding, Honghui, Tang, Hongxuan, Cao, Huanqi, Gao, Huazuo, Qu, Hui, Zeng, Hui, Huang, Jialiang, Li, Jiashi, Xu, Jiaxin, Hu, Jiewen, Chen, Jingchang, Xiang, Jingting, Yuan, Jingyang, Cheng, Jingyuan, Zhu, Jinhua, Ran, Jun, Jiang, Junguang, Qiu, Junjie, Li, Junlong, Song, Junxiao, Dong, Kai, Gao, Kaige, Guan, Kang, Huang, Kexin, Zhou, Kexing, Huang, Kezhao, Yu, Kuai, Wang, Lean, Zhang, Lecong, Wang, Lei, Zhao, Liang, Yin, Liangsheng, Guo, Lihua, Luo, Lingxiao, Ma, Linwang, Wang, Litong, Zhang, Liyue, Di, M. S., Xu, M. Y, Zhang, Mingchuan, Zhang, Minghua, Tang, Minghui, Zhou, Mingxu, Huang, Panpan, Cong, Peixin, Wang, Peiyi, Wang, Qiancheng, Zhu, Qihao, Li, Qingyang, Chen, Qinyu, Du, Qiushi, Xu, Ruiling, Ge, Ruiqi, Zhang, Ruisong, Pan, Ruizhe, Wang, Runji, Yin, Runqiu, Xu, Runxin, Shen, Ruomeng, Zhang, Ruoyu, Liu, S. H., Lu, Shanghao, Zhou, Shangyan, Chen, Shanhuang, Cai, Shaofei, Chen, Shaoyuan, Hu, Shengding, Liu, Shengyu, Hu, Shiqiang, Ma, Shirong, Wang, Shiyu, Yu, Shuiping, Zhou, Shunfeng, Pan, Shuting, Zhou, Songyang, Ni, Tao, Yun, Tao, Pei, Tian, Ye, Tian, Yue, Tianyuan, Zeng, Wangding, Liu, Wen, Liang, Wenfeng, Pang, Wenjie, Luo, Wenjing, Gao, Wenjun, Zhang, Wentao, Gao, Xi, Wang, Xiangwen, Bi, Xiao, Liu, Xiaodong, Wang, Xiaohan, Chen, Xiaokang, Zhang, Xiaokang, Nie, Xiaotao, Cheng, Xin, Liu, Xin, Xie, Xin, Liu, Xingchao, Yu, Xingkai, Li, Xingyou, Yang, Xinyu, Li, Xinyuan, Chen, Xu, Su, Xuecheng, Pan, Xuehai, Lin, Xuheng, Fu, Xuwei, Wang, Y. Q., Zhang, Yang, Xu, Yanhong, Ma, Yanru, Li, Yao, Li, Yao, Zhao, Yao, Sun, Yaofeng, Wang, Yaohui, Qian, Yi, Yu, Yi, Zhang, Yichao, Ding, Yifan, Shi, Yifan, Xiong, Yiliang, He, Ying, Zhou, Ying, Zhong, Yinmin, Piao, Yishi, Wang, Yisong, Chen, Yixiao, Tan, Yixuan, Wei, Yixuan, Ma, Yiyang, Liu, Yiyuan, Yang, Yonglun, Guo, Yongqiang, Wu, Yongtong, Wu, Yu, Cheng, Yuan, Ou, Yuan, Xu, Yuanfan, Wang, Yuduan, Gong, Yue, Wu, Yuhan, Zou, Yuheng, Li, Yukun, Xiong, Yunfan, Luo, Yuxiang, You, Yuxiang, Liu, Yuxuan, Zhou, Yuyang, Wu, Z. F., Ren, Z. Z., Zhao, Zehua, Ren, Zehui, Sha, Zhangli, Fu, Zhe, Xu, Zhean, Xie, Zhenda, Zhang, Zhengyan, Hao, Zhewen, Gou, Zhibin, Ma, Zhicheng, Yan, Zhigang, Shao, Zhihong, Huang, Zhixian, Wu, Zhiyu, Li, Zhuoshu, Zhang, Zhuping, Xu, Zian, Wang, Zihao, Gu, Zihui, Zhu, Zijia, Li, Zilin, Zhang, Zipeng, Xie, Ziwei, Gao, Ziyi, Pan, Zizheng, Yao, Zongqing, Feng, Bei, Li, Hui, Cai, J. L., Ni, Jiaqi, Xu, Lei, Li, Meng, Tian, Ning, Chen, R. J., Jin, R. L., Li, S. S., Zhou, Shuang, Sun, Tianyu, Li, X. Q., Jin, Xiangyue, Shen, Xiaojin, Chen, Xiaosha, Song, Xinnan, Zhou, Xinyi, Zhu, Y. X., Huang, Yanping, Li, Yaohui, Zheng, Yi, Zhu, Yuchen, Ma, Yunxian, Huang, Zhen, Xu, Zhipeng, Zhang, Zhongyu, Ji, Dongjie, Liang, Jian, Guo, Jianzhong, Chen, Jin, Xia, Leyi, Wang, Miaojun, Li, Mingming, Zhang, Peng, Chen, Ruyi, Sun, Shangmian, Wu, Shaoqing, Ye, Shengfeng, Wang, T., Xiao, W. L., An, Wei, Wang, Xianzu, Sun, Xiaowen, Wang, Xiaoxiang, Tang, Ying, Zha, Yukun, Zhang, Zekai, Ju, Zhe, Zhang, Zhen, Qu, Zihua

  15. Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments — published result table ↗

    Froger, Romain, Andrews, Pierre, Bettini, Matteo, Budhiraja, Amar, Cabral, Ricardo Silveira, Do, Virginie et al. · 2026

    All 24 authors

    Froger, Romain, Andrews, Pierre, Bettini, Matteo, Budhiraja, Amar, Cabral, Ricardo Silveira, Do, Virginie, Garreau, Emilien, Gaya, Jean-Baptiste, Laurençon, Hugo, Lecanu, Maxime, Malkan, Kunal, Mekala, Dheeraj, Ménard, Pierre, Bertran, Gerard Moreno-Torres, Piterbarg, Ulyana, Plekhanov, Mikhail, Rita, Mathieu, Rusakov, Andrey, Vorotilov, Vladislav, Wang, Mengjue, Yu, Ian, Benhalloum, Amine, Mialon, Grégoire, Scialom, Thomas

  16. TasteVal: Measuring the Experimental Research Taste of AI Systems Against Human Experts — published result table ↗

    Jaffe, Oliver, Sherburn, Dane · 2026

  17. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  18. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  19. FutureSearch Evals ↗

    FutureSearch

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

    EconBench project · 2026

  21. Ale Bench — methodology and published results ↗

    Epoch AI · 2026

  22. Algotune — methodology and published results ↗

    Epoch AI · 2026

  23. ARC-AGI-2 ↗

    Epoch AI

  24. Chess Puzzles ↗

    Epoch AI

  25. CritPt v1 ↗

    Epoch AI

  26. Earthborne Rangers (EBR-bench) ↗

    Epoch AI

  27. FrontierMath Tier 4 (v1, old) ↗

    Epoch AI

  28. FrontierMath Tier 4 (v2) ↗

    Epoch AI

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

    Epoch AI

  30. FrontierMath Tiers 1-3 (v2) ↗

    Epoch AI

  31. GDPval ↗

    Epoch AI

  32. GPQA Diamond ↗

    Epoch AI

  33. Humanity's Last Exam ↗

    Epoch AI

  34. MATH Level 5 ↗

    Epoch AI

  35. METR Time Horizons ↗

    Epoch AI

  36. Mystery Game Puzzles ↗

    Epoch AI

  37. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  38. Remote Labor Index ↗

    Epoch AI

  39. SimpleQA Verified ↗

    Epoch AI

  40. SWE-bench Verified ↗

    Epoch AI

  41. Terminal-Bench 2.0 ↗

    Epoch AI

  42. Capabilities & benchmarking — retained external evaluation records ↗

    Epoch AI · 2026

  43. AI benchmarks & capabilities ↗

    Epoch AI

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

    Fiction.live · 2025

  45. GeoBench — GeoGuessr model benchmark ↗

    GeoBench

  46. Writing benchmark — incorporating mandatory story elements ↗

    lechmazur

  47. GSO: Challenging Software Optimization Tasks for Evaluating SWE-Agents ↗

    Shetty, Manish, Jain, Naman, Liu, Jinjian, Kethanaboyina, Vijay, Sen, Koushik, Stoica, Ion · 2025

  48. WeirdML ↗

    Håvard Tveit Ihle

  49. Biomni-R0: Using RL to Hill-Climb Biomedical Reasoning Agents to Expert-Level ↗

    Ryan Li, Kexin Huang, Shiyi Cao, Yuanhao Qu, Jure Leskovec · 2025

  50. EnigmaEval ↗

    Scale Labs / SEAL

  51. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  52. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  53. AI Model Benchmarks — LM Council ↗

    LM Council

  54. Measuring AI Ability to Complete Long Software Tasks ↗

    METR

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

    METR · 2026

  56. Terminal-Bench 2.0 leaderboard ↗

    Terminal-Bench Team

  57. AIME — evaluation methodology and results ↗

    Vals AI · 2026

  58. CaseLaw — evaluation methodology and results ↗

    Vals AI · 2026

  59. CorpFin — evaluation methodology and results ↗

    Vals AI · 2026

  60. GPQA Diamond — evaluation methodology and results ↗

    Vals AI · 2026

  61. LiveCodeBench — evaluation methodology and results ↗

    Vals AI · 2026

  62. LegalBench — evaluation methodology and results ↗

    Vals AI · 2026

  63. MATH 500 — evaluation methodology and results ↗

    Vals AI · 2026

  64. MedCode — evaluation methodology and results ↗

    Vals AI · 2026

  65. MedQA — evaluation methodology and results ↗

    Vals AI · 2026

  66. MedScribe — evaluation methodology and results ↗

    Vals AI · 2026

  67. MGSM — evaluation methodology and results ↗

    Vals AI · 2026

  68. MMLU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  69. MMMU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  70. MortgageTax — evaluation methodology and results ↗

    Vals AI · 2026

  71. Poker Agent — evaluation methodology and results ↗

    Vals AI · 2026

  72. ProofBench — evaluation methodology and results ↗

    Vals AI · 2026

  73. SAGE — evaluation methodology and results ↗

    Vals AI · 2026

  74. SWE-bench Verified — evaluation methodology and results ↗

    Vals AI · 2026

  75. TaxEval — evaluation methodology and results ↗

    Vals AI · 2026

  76. Vibe Code Bench — evaluation methodology and results ↗

    Vals AI · 2026

Names used in source tables

GPT 5 · GPT 5 (Mini-SWE-agent) · GPT 5 (OpenHands) · GPT-5 · GPT-5 (High) · GPT-5 (high) · GPT-5 (low) · GPT-5 (medium) · GPT-5 (unknown thinking) · GPT-5 High · gpt-5 · gpt-5 (high) · gpt-5 (low) · gpt-5 (medium)