τ²-bench

τ²-bench Banking

Source

Artificial Analysis

Version

Not reported

Measurement date

Not reported

Metric

% · Higher is better

Evaluation scope

Tool-agent benchmark in the banking domain; observed 2026-10-19; result date not supplied

Exact score ties share a position. Positions do not establish statistical significance. Missing metadata is shown as unreported; matching imported metadata is not independent verification of the protocol.

1. #1Grok 4.6

   x-ai/grok-4.6-20260810

   50.7%

   View modelEvidence independent

2. #2Muse Spark 1.3

   meta/muse-spark-1.3-20260902

   50.5%

   View modelEvidence independent

3. #3GLM 5.3

   z-ai/glm-5.3-20260816

   50.3%

   View modelEvidence independent

4. #4Qwen3.8 27B

   qwen/qwen3.8-27b-20260814

   48%

   View modelEvidence independent

5. #5Qwen3.8 Max (0902)

   qwen/qwen3.8-max-20260902

   47.8%

   View modelEvidence independent

6. #6Claude Fable 5.1

   anthropic/claude-fable-5-1

   47.2%

   View modelEvidence independent

7. #6GLM 5.3 Flash

   z-ai/glm-5.3-flash-20260826

   47.2%

   View modelEvidence independent

8. #8Kimi K3

   moonshotai/kimi-k3-20260715

   46%

   View modelEvidence independent

9. #9Gemini 3.8 Flash

   google/gemini-3.8-flash-20260902

   44.9%

   View modelEvidence independent

10. #10GPT-5.6 Sol

    openai/gpt-5.6-sol-20260709

    44.3%

    View modelEvidence independent

11. #11Claude Opus 5

    anthropic/claude-opus-5

    42.1%

    View modelEvidence independent

12. #12GPT-6 Astra

    openai/gpt-6-astra-20260903

    41.4%

    View modelEvidence independent

13. #13GPT-5.6 Terra

    openai/gpt-5.6-terra-20260709

    40.2%

    View modelEvidence independent

14. #14DeepSeek V4 Pro 0813

    deepseek/deepseek-v4-pro-20260813

    39.6%

    View modelEvidence independent

15. #15Claude Fable 5

    anthropic/claude-fable-5

    38.1%

    View modelEvidence independent

16. #16Muse Spark 1.2

    meta/muse-spark-1.2-20260805

    34.8%

    View modelEvidence independent

17. #17Gemini 3.7 Flash

    google/gemini-3.7-flash-20260813

    32.8%

    View modelEvidence independent

18. #18GPT-5.6 Luna

    openai/gpt-5.6-luna-20260709

    31.1%

    View modelEvidence independent

19. #19Inkling

    thinkingmachines/inkling-20260715

    29.1%

    View modelEvidence independent

20. #20Hy3

    tencent/hy3-20260706

    22.9%

    View modelEvidence independent

21. #20Hy3 preview

    tencent/hy3-preview-20260421

    22.9%

    View modelEvidence independent

22. #22Gemini 3.5 Flash Lite

    google/gemini-3.5-flash-lite-20260721

    17.5%

    View modelEvidence independent

23. #23MiniMax M3

    minimax/minimax-m3-20260531

    15.3%

    View modelEvidence independent

24. #24Mistral Medium 3

    mistralai/mistral-medium-3

    15.1%

    View modelEvidence independent

25. #24Mistral Medium 3.1

    mistralai/mistral-medium-3.1

    15.1%

    View modelEvidence independent

26. #24Mistral Medium 3.5

    mistralai/mistral-medium-3.5-20260430

    15.1%

    View modelEvidence independent

27. #27Nemotron 3 Ultra

    nvidia/nemotron-3-ultra-550b-a55b-20260604

    14.2%

    View modelEvidence independent

28. #28gpt-oss-120b

    openai/gpt-oss-120b

    12.8%

    View modelEvidence independent

29. #29MiMo-V2.5-Pro

    xiaomi/mimo-v2.5-pro-20260422

    9.9%

    View modelEvidence independent

30. #30Nemotron 3.5 Lightning

    nvidia/nemotron-3.5-lightning-20260807

    8.9%

    View modelEvidence independent

31. #31MiMo-V2.5

    xiaomi/mimo-v2.5-20260422

    8.7%

    View modelEvidence independent

32. #32Command A

    cohere/command-a-03-2025

    6%

    View modelEvidence independent

33. #32Command A+

    cohere/command-a-plus-05-2026

    6%

    View modelEvidence independent