Acropolis

EMB

EMB, a component of the Vals Index.

Upstream source
FinanceAgentic
Independent. Run by an operator independent of the model vendors.Vals AI benchmarks by Vals AIVersion 1.1Data as of 17 Nov 2026

No filters. Showing 38 of 38 rows.

  1. 1
    AnthropicDossier

    high

    Accuracy

    Fable 5.2, Accuracy: 77.4% (scale 0.0 to 100.0), 95% confidence interval 73.1 to 81.777.4%(73.1–81.7)
    Duration per test
    979.1 s
    Cost per test
    $6.67
  2. 2
    GoogleDossier

    high

    Tied #1

    Accuracy

    Gemini 4 Carbon, Accuracy: 76.5% (scale 0.0 to 100.0), 95% confidence interval 72.1 to 80.876.5%(72.1–80.8)
    Duration per test
    1,648.6 s
    Cost per test
    $5.76
  3. 3
    AnthropicDossier

    max

    Tied #1

    Accuracy

    Sonnet 5.5, Accuracy: 75.9% (scale 0.0 to 100.0), 95% confidence interval 71.1 to 80.775.9%(71.1–80.7)
    Duration per test
    2,205 s
    Cost per test
    $6.53
  4. 4
    AnthropicDossier

    max

    Tied #1

    Accuracy

    Opus 5.5, Accuracy: 75.8% (scale 0.0 to 100.0), 95% confidence interval 71.1 to 80.675.8%(71.1–80.6)
    Duration per test
    2,106.4 s
    Cost per test
    $9.33
  5. 5
    GoogleDossier

    xhigh

    Tied #1

    Accuracy

    Gemini 4 Hydrogen, Accuracy: 75.8% (scale 0.0 to 100.0), 95% confidence interval 71.8 to 79.875.8%(71.8–79.8)
    Duration per test
    662.9 s
    Cost per test
    $3.17
  6. 6

    max

    Tied #1

    Accuracy

    Muse Wonder Full 1.0, Accuracy: 75% (scale 0.0 to 100.0), 95% confidence interval 70.8 to 79.275%(70.8–79.2)
    Duration per test
    1,684.5 s
    Cost per test
    $4.58
  7. 7
    GoogleDossier

    high

    Tied #1

    Accuracy

    Gemini 4 Argon, Accuracy: 74.8% (scale 0.0 to 100.0), 95% confidence interval 69.9 to 79.774.8%(69.9–79.7)
    Duration per test
    2,305.8 s
    Cost per test
    $11.80
  8. 8
    OpenAIDossier

    max

    Tied #1

    Accuracy

    GPT-6.1 Terra, Accuracy: 73.3% (scale 0.0 to 100.0), 95% confidence interval 68.9 to 77.773.3%(68.9–77.7)
    Duration per test
    756.9 s
    Cost per test
    $0.56
  9. 9
    GoogleDossier

    high

    Tied #1

    Accuracy

    Gemini 3.8 Flash, Accuracy: 72.9% (scale 0.0 to 100.0), 95% confidence interval 68.1 to 77.672.9%(68.1–77.6)
    Duration per test
    769.2 s
    Cost per test
    $8.24
  10. 10
    OpenAIDossier

    max

    Tied #1

    Accuracy

    GPT-6 Astra, Accuracy: 72.7% (scale 0.0 to 100.0), 95% confidence interval 68.2 to 77.272.7%(68.2–77.2)
    Duration per test
    804.3 s
    Cost per test
    $5.83
  11. 11

    max

    Tied #1

    Accuracy

    Muse Wonder 1.0, Accuracy: 72.5% (scale 0.0 to 100.0), 95% confidence interval 67.8 to 77.272.5%(67.8–77.2)
    Duration per test
    1,034.4 s
    Cost per test
    $5.60
  12. 12

    max

    Tied #1

    Accuracy

    Muse Ember, Accuracy: 72.2% (scale 0.0 to 100.0), 95% confidence interval 67.6 to 76.772.2%(67.6–76.7)
    Duration per test
    289.8 s
    Cost per test
    $0.091
  13. 13
    AlibabaDossier
    Tied #1

    Accuracy

    Qwen3.8 Max Internal, Accuracy: 72% (scale 0.0 to 100.0), 95% confidence interval 67.3 to 76.772%(67.3–76.7)
    Duration per test
    3,193.4 s
    Cost per test
    $4.43
  14. 14

    xhigh

    Tied #1

    Accuracy

    Grok 4.7, Accuracy: 71.5% (scale 0.0 to 100.0), 95% confidence interval 66.9 to 76.171.5%(66.9–76.1)
    Duration per test
    2,097 s
    Cost per test
    $6.48
  15. 15
    Moonshot AIDossier

    max

    Tied #1

    Accuracy

    Kimi K3, Accuracy: 71.3% (scale 0.0 to 100.0), 95% confidence interval 66.8 to 75.771.3%(66.8–75.7)
    Duration per test
    741.4 s
    Cost per test
    $3.07
  16. 16
    OpenAIDossier

    max

    Tied #1

    Accuracy

    GPT-6.1 Sol, Accuracy: 71.1% (scale 0.0 to 100.0), 95% confidence interval 65.7 to 76.571.1%(65.7–76.5)
    Duration per test
    1,107.8 s
    Cost per test
    $1.14
  17. 17
    AlibabaDossier

    xhigh

    Accuracy

    Qwen4 Max, Accuracy: 70.8% (scale 0.0 to 100.0), 95% confidence interval 66.2 to 75.570.8%(66.2–75.5)
    Duration per test
    970.6 s
    Cost per test
    $1.00
  18. 18
    Moonshot AIDossier

    max

    Accuracy

    Kimi K3.1 Pro, Accuracy: 70% (scale 0.0 to 100.0), 95% confidence interval 64.8 to 75.270%(64.8–75.2)
    Duration per test
    1,283.3 s
    Cost per test
    $1.59
  19. 19
    OpenAIDossier

    max

    Accuracy

    GPT-6 Luna, Accuracy: 70% (scale 0.0 to 100.0), 95% confidence interval 65.0 to 74.970%(65.0–74.9)
    Duration per test
    1,041.1 s
    Cost per test
    $0.14
  20. 20
    StepFunDossier

    Accuracy

    Step 5 Preview, Accuracy: 69.6% (scale 0.0 to 100.0), 95% confidence interval 64.7 to 74.469.6%(64.7–74.4)
    Duration per test
    2,615.1 s
    Cost per test
    $1.33
  21. 21
    Thinking MachinesDossier

    Accuracy

    Inkling 2 Small, Accuracy: 69.5% (scale 0.0 to 100.0), 95% confidence interval 64.6 to 74.469.5%(64.6–74.4)
    Duration per test
    833.9 s
    Cost per test
    $0.53
  22. 22
    DeepSeekDossier

    max

    Accuracy

    DeepSeek V4.1 Flash 1017, Accuracy: 69.2% (scale 0.0 to 100.0), 95% confidence interval 64.5 to 73.969.2%(64.5–73.9)
    Duration per test
    1,971.7 s
    Cost per test
    $0.33
  23. 23

    max

    Accuracy

    GLM-5.3, Accuracy: 69.1% (scale 0.0 to 100.0), 95% confidence interval 63.9 to 74.369.1%(63.9–74.3)
    Duration per test
    2,172.2 s
    Cost per test
    $3.79
  24. 24
    TencentDossier

    Accuracy

    Hy4 Preview, Accuracy: 68.9% (scale 0.0 to 100.0), 95% confidence interval 64.2 to 73.768.9%(64.2–73.7)
    Duration per test
    1,839.4 s
    Cost per test
    $0.79
  25. 25
    XiaomiDossier

    Accuracy

    MiMo-V2.6-Flash, Accuracy: 68.9% (scale 0.0 to 100.0), 95% confidence interval 64.4 to 73.368.9%(64.4–73.3)
    Duration per test
    1,521.9 s
    Cost per test
    $0.11
  26. 26
    MistralDossier

    high

    Accuracy

    Mistral Large 4 OSS, Accuracy: 68.7% (scale 0.0 to 100.0), 95% confidence interval 63.9 to 73.468.7%(63.9–73.4)
    Duration per test
    1,387.4 s
    Cost per test
    $0.83
  27. 27
    MiniMaxDossier

    max

    Accuracy

    MiniMax M3.1 Flash, Accuracy: 67.8% (scale 0.0 to 100.0), 95% confidence interval 63.4 to 72.367.8%(63.4–72.3)
    Duration per test
    1,847.8 s
    Cost per test
    $0.27
  28. 28
    XiaomiDossier

    Accuracy

    MiMo-V2.6-Pro, Accuracy: 67.3% (scale 0.0 to 100.0), 95% confidence interval 61.7 to 73.067.3%(61.7–73.0)
    Duration per test
    3,313.9 s
    Cost per test
    $0.33
  29. 29

    max

    Accuracy

    GLM-5.3 Flash, Accuracy: 66.3% (scale 0.0 to 100.0), 95% confidence interval 61.0 to 71.566.3%(61.0–71.5)
    Duration per test
    4,495.2 s
    Cost per test
    $1.43
  30. 30
    TencentDossier

    Accuracy

    Hy4, Accuracy: 66.2% (scale 0.0 to 100.0), 95% confidence interval 61.0 to 71.466.2%(61.0–71.4)
    Duration per test
    1,058.9 s
    Cost per test
    $0.62
  31. 31
    AnthropicDossier

    max

    Accuracy

    Haiku 5.5, Accuracy: 65.5% (scale 0.0 to 100.0), 95% confidence interval 58.2 to 72.765.5%(58.2–72.7)
    Duration per test
    1,195.5 s
    Cost per test
    $1.36
  32. 32
    DeepSeekDossier

    max

    Accuracy

    DeepSeek V4 Pro 0813, Accuracy: 65.1% (scale 0.0 to 100.0), 95% confidence interval 60.5 to 69.765.1%(60.5–69.7)
    Duration per test
    2,247.2 s
    Cost per test
    $1.08
  33. 33
    ReflectionDossier

    Accuracy

    Beam 1.0, Accuracy: 64.7% (scale 0.0 to 100.0), 95% confidence interval 60.3 to 69.264.7%(60.3–69.2)
    Duration per test
    1,204.3 s
    Cost per test
    Not tested
  34. 34

    xhigh

    Accuracy

    Muse Spark Open, Accuracy: 62% (scale 0.0 to 100.0), 95% confidence interval 56.7 to 67.362%(56.7–67.3)
    Duration per test
    5,078.6 s
    Cost per test
    $5.35
  35. 35
    InceptionDossier

    high

    Accuracy

    Mercury 3, Accuracy: 57.9% (scale 0.0 to 100.0), 95% confidence interval 52.2 to 63.657.9%(52.2–63.6)
    Duration per test
    531.3 s
    Cost per test
    $0.19
  36. 36
    NVIDIADossier

    high

    Accuracy

    Nemotron 3.5 Super VL, Accuracy: 52.8% (scale 0.0 to 100.0), 95% confidence interval 47.3 to 58.252.8%(47.3–58.2)
    Duration per test
    816.7 s
    Cost per test
    $1.29
  37. 37
    Thinking MachinesDossier

    Accuracy

    Inkling, Accuracy: 41% (scale 0.0 to 100.0), 95% confidence interval 35.1 to 46.841%(35.1–46.8)
    Duration per test
    1,581.8 s
    Cost per test
    $1.67
  38. 38
    Thinking MachinesDossier

    Accuracy

    Inkling Small, Accuracy: 40.8% (scale 0.0 to 100.0), 95% confidence interval 34.5 to 47.140.8%(34.5–47.1)
    Duration per test
    1,173.6 s
    Cost per test
    $0.25