Acropolis

LMArena WebDev

Arena human-vote leaderboard: rating with confidence interval, vote count and rank. One row per model and category.

Coding
Independent. Run by an operator independent of the model vendors.LMArena by ArenaVersion 1Data as of 17 Nov 2026

No filters. Showing 156 of 156 rows.

  1. 1
    AnthropicDossier

    xhigh

    Arena rating

    Fable 5.2, Arena rating: 1,826 (scale 1000.0 to 1826.0), 95% confidence interval 1807.2 to 1844.81,826(1807.2–1844.8)
  2. 2
    AnthropicDossier

    max

    Tied #1

    Arena rating

    Opus 5.5, Arena rating: 1,816 (scale 1000.0 to 1826.0), 95% confidence interval 1804.0 to 1827.11,816(1804.0–1827.1)
  3. 3
    AnthropicDossier

    xhigh

    Arena rating

    Sonnet 5.5, Arena rating: 1,772 (scale 1000.0 to 1826.0), 95% confidence interval 1762.5 to 1781.91,772(1762.5–1781.9)
  4. 4
    GoogleDossier

    high

    Arena rating

    Gemini 4 Carbon, Arena rating: 1,768 (scale 1000.0 to 1826.0), 95% confidence interval 1753.0 to 1783.71,768(1753.0–1783.7)
  5. 5
    OpenAIDossier

    max

    Arena rating

    GPT-6 Astra, Arena rating: 1,763 (scale 1000.0 to 1826.0), 95% confidence interval 1754.4 to 1771.31,763(1754.4–1771.3)
  6. 6
    AnthropicDossier

    max

    Arena rating

    Fable 5.1, Arena rating: 1,743 (scale 1000.0 to 1826.0), 95% confidence interval 1735.1 to 1750.81,743(1735.1–1750.8)
  7. 7
    OpenAIDossier

    max

    Arena rating

    GPT-6.1 Sol, Arena rating: 1,742 (scale 1000.0 to 1826.0), 95% confidence interval 1732.7 to 1752.01,742(1732.7–1752.0)
  8. 8

    max

    Arena rating

    Muse Wonder Full 1.0, Arena rating: 1,724 (scale 1000.0 to 1826.0), 95% confidence interval 1709.4 to 1738.31,724(1709.4–1738.3)
  9. 9
    AnthropicDossier

    high

    Arena rating

    Sonnet 5.5, Arena rating: 1,716 (scale 1000.0 to 1826.0), 95% confidence interval 1707.5 to 1724.11,716(1707.5–1724.1)
  10. 10
    GoogleDossier

    high

    Arena rating

    Gemini 4 Hydrogen, Arena rating: 1,704 (scale 1000.0 to 1826.0), 95% confidence interval 1689.7 to 1718.11,704(1689.7–1718.1)
  11. 11
    AnthropicDossier

    max

    Arena rating

    Opus 5, Arena rating: 1,691 (scale 1000.0 to 1826.0), 95% confidence interval 1685.0 to 1697.21,691(1685.0–1697.2)
  12. 12
    Moonshot AIDossier

    max

    Arena rating

    Kimi K3.1 Pro, Arena rating: 1,686 (scale 1000.0 to 1826.0), 95% confidence interval 1676.7 to 1695.71,686(1676.7–1695.7)
  13. 13
    OpenAIDossier

    max

    Arena rating

    GPT-6.1 Terra, Arena rating: 1,681 (scale 1000.0 to 1826.0), 95% confidence interval 1670.4 to 1691.01,681(1670.4–1691.0)
  14. 14
    GoogleDossier

    high

    Arena rating

    Gemini 4 Argon, Arena rating: 1,679 (scale 1000.0 to 1826.0), 95% confidence interval 1669.6 to 1688.61,679(1669.6–1688.6)
  15. 15
    AlibabaDossier

    Arena rating

    Qwen3.8 Max (0902), Arena rating: 1,676 (scale 1000.0 to 1826.0), 95% confidence interval 1669.4 to 1681.71,676(1669.4–1681.7)
  16. 16
    OpenAIDossier

    max

    Arena rating

    GPT-6 Sol, Arena rating: 1,673 (scale 1000.0 to 1826.0), 95% confidence interval 1664.6 to 1681.01,673(1664.6–1681.0)
  17. 17
    AlibabaDossier

    Arena rating

    Qwen3.8-Max, Arena rating: 1,670 (scale 1000.0 to 1826.0), 95% confidence interval 1658.4 to 1682.61,670(1658.4–1682.6)
  18. 18
    DeepSeekDossier

    max

    Arena rating

    DeepSeek V4.1 Flash 1017, Arena rating: 1,661 (scale 1000.0 to 1826.0), 95% confidence interval 1654.7 to 1668.11,661(1654.7–1668.1)
  19. 19
    Moonshot AIDossier

    max

    Arena rating

    Kimi K3, Arena rating: 1,658 (scale 1000.0 to 1826.0), 95% confidence interval 1652.1 to 1663.41,658(1652.1–1663.4)
  20. 20
    AnthropicDossier

    high

    Arena rating

    Opus 5, Arena rating: 1,657 (scale 1000.0 to 1826.0), 95% confidence interval 1651.2 to 1661.81,657(1651.2–1661.8)
  21. 21

    max

    Arena rating

    Muse Spark 1.3, Arena rating: 1,655 (scale 1000.0 to 1826.0), 95% confidence interval 1648.2 to 1662.01,655(1648.2–1662.0)
  22. 22

    max

    Arena rating

    Muse Wonder 1.0, Arena rating: 1,647 (scale 1000.0 to 1826.0), 95% confidence interval 1638.2 to 1656.51,647(1638.2–1656.5)
  23. 23
    Thinking MachinesDossier

    Arena rating

    Inkling 2 Small, Arena rating: 1,645 (scale 1000.0 to 1826.0), 95% confidence interval 1626.1 to 1663.31,645(1626.1–1663.3)
  24. 24

    xhigh

    Arena rating

    Grok 4.7, Arena rating: 1,640 (scale 1000.0 to 1826.0), 95% confidence interval 1632.3 to 1647.91,640(1632.3–1647.9)
  25. 25
    XiaomiDossier

    Arena rating

    MiMo-V2.6-Flash, Arena rating: 1,639 (scale 1000.0 to 1826.0), 95% confidence interval 1630.9 to 1646.11,639(1630.9–1646.1)
  26. 26
    TencentDossier

    Arena rating

    Hy4 Preview, Arena rating: 1,632 (scale 1000.0 to 1826.0), 95% confidence interval 1621.8 to 1641.41,632(1621.8–1641.4)
  27. 27
    AnthropicDossier

    high

    Arena rating

    Fable 5, Arena rating: 1,626 (scale 1000.0 to 1826.0), 95% confidence interval 1619.7 to 1631.61,626(1619.7–1631.6)
  28. 28
    AlibabaDossier

    Arena rating

    Qwen3.8 Flash Next, Arena rating: 1,625 (scale 1000.0 to 1826.0), 95% confidence interval 1617.5 to 1631.51,625(1617.5–1631.5)
  29. 29
    XiaomiDossier

    Arena rating

    MiMo-V2.6-Pro, Arena rating: 1,624 (scale 1000.0 to 1826.0), 95% confidence interval 1616.3 to 1631.41,624(1616.3–1631.4)
  30. 30

    max

    Arena rating

    GLM-5.3, Arena rating: 1,622 (scale 1000.0 to 1826.0), 95% confidence interval 1615.5 to 1629.11,622(1615.5–1629.1)
  31. 31

    xhigh

    Arena rating

    Muse Spark 1.3, Arena rating: 1,621 (scale 1000.0 to 1826.0), 95% confidence interval 1614.5 to 1628.31,621(1614.5–1628.3)
  32. 32
    DeepSeekDossier

    max

    Arena rating

    DeepSeek V4.1 Flash, Arena rating: 1,621 (scale 1000.0 to 1826.0), 95% confidence interval 1609.6 to 1631.91,621(1609.6–1631.9)
  33. 33

    high

    Arena rating

    Grok 4.6, Arena rating: 1,618 (scale 1000.0 to 1826.0), 95% confidence interval 1611.2 to 1625.11,618(1611.2–1625.1)
  34. 34
    OpenAIDossier

    xhigh · Codex CLI

    Arena rating

    GPT-5.6 Sol, Arena rating: 1,616 (scale 1000.0 to 1826.0), 95% confidence interval 1610.0 to 1621.51,616(1610.0–1621.5)
  35. 35
    UnisoundDossier

    Arena rating

    U2 Flash, Arena rating: 1,612 (scale 1000.0 to 1826.0), 95% confidence interval 1595.6 to 1628.91,612(1595.6–1628.9)
  36. 36

    Arena rating

    GLM-5.3 Flash, Arena rating: 1,609 (scale 1000.0 to 1826.0), 95% confidence interval 1603.0 to 1615.21,609(1603.0–1615.2)
  37. 37

    max

    Arena rating

    Muse Ember, Arena rating: 1,608 (scale 1000.0 to 1826.0), 95% confidence interval 1591.1 to 1624.81,608(1591.1–1624.8)
  38. 38

    max

    Arena rating

    GLM-5.2, Arena rating: 1,608 (scale 1000.0 to 1826.0), 95% confidence interval 1601.7 to 1613.51,608(1601.7–1613.5)
  39. 39
    AnthropicDossier

    high

    Arena rating

    Haiku 5.5, Arena rating: 1,593 (scale 1000.0 to 1826.0), 95% confidence interval 1585.0 to 1601.41,593(1585.0–1601.4)
  40. 40
    AlibabaDossier

    Arena rating

    Qwen3.8-27B, Arena rating: 1,590 (scale 1000.0 to 1826.0), 95% confidence interval 1583.9 to 1596.01,590(1583.9–1596.0)
  41. 41
    GoogleDossier

    high

    Arena rating

    Gemini 3.7 Flash, Arena rating: 1,588 (scale 1000.0 to 1826.0), 95% confidence interval 1581.9 to 1594.81,588(1581.9–1594.8)
  42. 42
    GoogleDossier

    high

    Arena rating

    Gemini 3.8 Flash, Arena rating: 1,583 (scale 1000.0 to 1826.0), 95% confidence interval 1576.4 to 1589.51,583(1576.4–1589.5)
  43. 43
    MiniMaxDossier

    Arena rating

    MiniMax M3.1 Flash, Arena rating: 1,582 (scale 1000.0 to 1826.0), 95% confidence interval 1572.6 to 1591.41,582(1572.6–1591.4)
  44. 44
    DeepSeekDossier

    high

    Arena rating

    DeepSeek V4-Flash, Arena rating: 1,582 (scale 1000.0 to 1826.0), 95% confidence interval 1572.2 to 1591.01,582(1572.2–1591.0)
  45. 45
    DeepSeekDossier

    high

    Arena rating

    DeepSeek V4 Pro 0813, Arena rating: 1,581 (scale 1000.0 to 1826.0), 95% confidence interval 1571.3 to 1590.41,581(1571.3–1590.4)
  46. 46
    OpenAIDossier

    max

    Arena rating

    GPT-6 Luna, Arena rating: 1,577 (scale 1000.0 to 1826.0), 95% confidence interval 1570.7 to 1582.51,577(1570.7–1582.5)
  47. 47
    StepFunDossier

    high

    Arena rating

    Step 5 Preview, Arena rating: 1,572 (scale 1000.0 to 1826.0), 95% confidence interval 1563.8 to 1579.71,572(1563.8–1579.7)
  48. 48
    MistralDossier

    Arena rating

    Mistral Large 4, Arena rating: 1,566 (scale 1000.0 to 1826.0), 95% confidence interval 1552.3 to 1580.41,566(1552.3–1580.4)
  49. 49
    AnthropicDossier

    high

    Arena rating

    Opus 4.7, Arena rating: 1,557 (scale 1000.0 to 1826.0), 95% confidence interval 1551.3 to 1562.81,557(1551.3–1562.8)
  50. 50
    AnthropicDossier

    high

    Arena rating

    Opus 4.8, Arena rating: 1,555 (scale 1000.0 to 1826.0), 95% confidence interval 1549.5 to 1561.11,555(1549.5–1561.1)
  51. 51
    AnthropicDossier

    Arena rating

    Opus 4.7, Arena rating: 1,555 (scale 1000.0 to 1826.0), 95% confidence interval 1549.6 to 1560.91,555(1549.6–1560.9)
  52. 52

    Arena rating

    Grok 4.5, Arena rating: 1,552 (scale 1000.0 to 1826.0), 95% confidence interval 1545.7 to 1559.21,552(1545.7–1559.2)
  53. 53
    AnthropicDossier

    high

    Arena rating

    Opus 4.6, Arena rating: 1,547 (scale 1000.0 to 1826.0), 95% confidence interval 1541.3 to 1552.31,547(1541.3–1552.3)
  54. 54

    Arena rating

    Muse Spark 1.1, Arena rating: 1,542 (scale 1000.0 to 1826.0), 95% confidence interval 1533.9 to 1549.21,542(1533.9–1549.2)
  55. 55
    AnthropicDossier

    high

    Arena rating

    Sonnet 5, Arena rating: 1,539 (scale 1000.0 to 1826.0), 95% confidence interval 1533.3 to 1544.91,539(1533.3–1544.9)
  56. 56
    GoogleDossier

    high

    Arena rating

    Gemini 3.6 Flash, Arena rating: 1,538 (scale 1000.0 to 1826.0), 95% confidence interval 1530.0 to 1545.11,538(1530.0–1545.1)
  57. 57
    AnthropicDossier

    Arena rating

    Opus 4.6, Arena rating: 1,537 (scale 1000.0 to 1826.0), 95% confidence interval 1531.4 to 1542.11,537(1531.4–1542.1)
  58. 58
    AnthropicDossier

    Arena rating

    Opus 4.8, Arena rating: 1,536 (scale 1000.0 to 1826.0), 95% confidence interval 1530.7 to 1542.21,536(1530.7–1542.2)
  59. 59

    xhigh

    Arena rating

    Muse Spark 1.2, Arena rating: 1,534 (scale 1000.0 to 1826.0), 95% confidence interval 1520.1 to 1547.81,534(1520.1–1547.8)
  60. 60

    max

    Arena rating

    Muse Spark Open, Arena rating: 1,527 (scale 1000.0 to 1826.0), 95% confidence interval 1512.7 to 1542.21,527(1512.7–1542.2)
  61. 61
    AnthropicDossier

    Arena rating

    Sonnet 4.6, Arena rating: 1,522 (scale 1000.0 to 1826.0), 95% confidence interval 1517.1 to 1527.51,522(1517.1–1527.5)
  62. 62
    OpenAIDossier

    xhigh · Codex CLI

    Arena rating

    GPT-5.6 Terra, Arena rating: 1,520 (scale 1000.0 to 1826.0), 95% confidence interval 1514.2 to 1526.31,520(1514.2–1526.3)
  63. 63
    ByteDanceDossier

    Arena rating

    Seed 2.1 Pro Preview, Arena rating: 1,519 (scale 1000.0 to 1826.0), 95% confidence interval 1513.8 to 1524.91,519(1513.8–1524.9)
  64. 64
    OpenAIDossier

    xhigh · Codex CLI

    Arena rating

    GPT-5.6 Luna, Arena rating: 1,519 (scale 1000.0 to 1826.0), 95% confidence interval 1512.6 to 1525.51,519(1512.6–1525.5)
  65. 65
    AlibabaDossier

    Arena rating

    Qwen3.7 Max, Arena rating: 1,515 (scale 1000.0 to 1826.0), 95% confidence interval 1507.8 to 1522.61,515(1507.8–1522.6)
  66. 66
    OpenAIDossier

    xhigh · Codex CLI

    Arena rating

    GPT-5.5, Arena rating: 1,512 (scale 1000.0 to 1826.0), 95% confidence interval 1506.1 to 1518.11,512(1506.1–1518.1)
  67. 67
    Moonshot AIDossier

    Arena rating

    Kimi K2.6, Arena rating: 1,510 (scale 1000.0 to 1826.0), 95% confidence interval 1502.2 to 1516.81,510(1502.2–1516.8)
  68. 68

    Arena rating

    GLM-5.1, Arena rating: 1,508 (scale 1000.0 to 1826.0), 95% confidence interval 1501.2 to 1514.81,508(1501.2–1514.8)
  69. 69
    TencentDossier

    Arena rating

    Hy3, Arena rating: 1,506 (scale 1000.0 to 1826.0), 95% confidence interval 1498.0 to 1513.71,506(1498.0–1513.7)
  70. 70
    GoogleDossier

    high

    Arena rating

    Gemini 3.5 Flash, Arena rating: 1,500 (scale 1000.0 to 1826.0), 95% confidence interval 1492.6 to 1507.31,500(1492.6–1507.3)
  71. 71
    AnthropicDossier

    high

    Arena rating

    Opus 4.5, Arena rating: 1,493 (scale 1000.0 to 1826.0), 95% confidence interval 1485.1 to 1501.81,493(1485.1–1501.8)
  72. 72
    GoogleDossier

    medium

    Arena rating

    Gemini 3.5 Flash, Arena rating: 1,490 (scale 1000.0 to 1826.0), 95% confidence interval 1483.3 to 1496.71,490(1483.3–1496.7)
  73. 73
    OpenAIDossier

    high · Codex CLI

    Arena rating

    GPT-5.5, Arena rating: 1,487 (scale 1000.0 to 1826.0), 95% confidence interval 1481.2 to 1493.21,487(1481.2–1493.2)
  74. 74
    SynlanDossier

    Arena rating

    Spark-X2.6, Arena rating: 1,485 (scale 1000.0 to 1826.0), 95% confidence interval 1465.6 to 1504.21,485(1465.6–1504.2)
  75. 75
    MiniMaxDossier

    Arena rating

    MiniMax M3, Arena rating: 1,484 (scale 1000.0 to 1826.0), 95% confidence interval 1478.6 to 1489.51,484(1478.6–1489.5)
  76. 76
    AlibabaDossier

    Arena rating

    Qwen3.6 Max Preview, Arena rating: 1,482 (scale 1000.0 to 1826.0), 95% confidence interval 1469.8 to 1494.91,482(1469.8–1494.9)
  77. 77
    XiaomiDossier

    Arena rating

    MiMo-V2.5-Pro, Arena rating: 1,479 (scale 1000.0 to 1826.0), 95% confidence interval 1473.9 to 1484.51,479(1473.9–1484.5)
  78. 78
    Moonshot AIDossier

    Arena rating

    Kimi K2.7 Code, Arena rating: 1,474 (scale 1000.0 to 1826.0), 95% confidence interval 1465.0 to 1483.91,474(1465.0–1483.9)
  79. 79
    AnthropicDossier

    Arena rating

    Opus 4.5, Arena rating: 1,469 (scale 1000.0 to 1826.0), 95% confidence interval 1461.6 to 1476.11,469(1461.6–1476.1)
  80. 80
    DeepSeekDossier

    high

    Arena rating

    DeepSeek V4 Pro Preview, Arena rating: 1,464 (scale 1000.0 to 1826.0), 95% confidence interval 1457.2 to 1469.91,464(1457.2–1469.9)
  81. 81
    OpenAIDossier

    high · Codex CLI

    Arena rating

    GPT-5.4, Arena rating: 1,462 (scale 1000.0 to 1826.0), 95% confidence interval 1442.6 to 1480.51,462(1442.6–1480.5)
  82. 82
    AlibabaDossier

    Arena rating

    Qwen3.6 Plus, Arena rating: 1,461 (scale 1000.0 to 1826.0), 95% confidence interval 1455.9 to 1467.11,461(1455.9–1467.1)
  83. 83
    OpenAIDossier

    Codex CLI

    Arena rating

    GPT-5.5, Arena rating: 1,456 (scale 1000.0 to 1826.0), 95% confidence interval 1449.6 to 1461.51,456(1449.6–1461.5)
  84. 84
    DeepSeekDossier

    Arena rating

    DeepSeek V4-Pro, Arena rating: 1,449 (scale 1000.0 to 1826.0), 95% confidence interval 1442.1 to 1454.91,449(1442.1–1454.9)
  85. 85
    GoogleDossier

    Arena rating

    Gemini 3.1 Pro Preview, Arena rating: 1,446 (scale 1000.0 to 1826.0), 95% confidence interval 1441.3 to 1451.51,446(1441.3–1451.5)
  86. 86
    OpenAIDossier

    medium · Codex CLI

    Arena rating

    GPT-5.4, Arena rating: 1,442 (scale 1000.0 to 1826.0), 95% confidence interval 1423.4 to 1460.21,442(1423.4–1460.2)
  87. 87
    GoogleDossier

    Arena rating

    Gemini 3 Pro Preview, Arena rating: 1,441 (scale 1000.0 to 1826.0), 95% confidence interval 1432.3 to 1449.31,441(1432.3–1449.3)
  88. 88
    GoogleDossier

    Arena rating

    Gemini 3 Flash Preview, Arena rating: 1,440 (scale 1000.0 to 1826.0), 95% confidence interval 1431.2 to 1448.61,440(1431.2–1448.6)
  89. 89
    XiaomiDossier

    Arena rating

    MiMo-V2.5, Arena rating: 1,438 (scale 1000.0 to 1826.0), 95% confidence interval 1431.9 to 1444.01,438(1431.9–1444.0)
  90. 90
    Moonshot AIDossier

    thinking

    Arena rating

    Kimi K2.5, Arena rating: 1,437 (scale 1000.0 to 1826.0), 95% confidence interval 1431.3 to 1441.81,437(1431.3–1441.8)
  91. 91
    ReflectionDossier

    Arena rating

    Beam 1.0, Arena rating: 1,436 (scale 1000.0 to 1826.0), 95% confidence interval 1422.6 to 1449.91,436(1422.6–1449.9)
  92. 92

    Arena rating

    GLM-4.7, Arena rating: 1,436 (scale 1000.0 to 1826.0), 95% confidence interval 1423.7 to 1447.81,436(1423.7–1447.8)
  93. 93

    Arena rating

    GLM-5, Arena rating: 1,435 (scale 1000.0 to 1826.0), 95% confidence interval 1426.8 to 1443.21,435(1426.8–1443.2)
  94. 94
    XiaomiDossier

    Arena rating

    MiMo-V2-Pro, Arena rating: 1,434 (scale 1000.0 to 1826.0), 95% confidence interval 1426.1 to 1442.51,434(1426.1–1442.5)
  95. 95
    GoogleDossier

    Arena rating

    Gemini 3.5 Flash-Lite, Arena rating: 1,432 (scale 1000.0 to 1826.0), 95% confidence interval 1390.7 to 1473.41,432(1390.7–1473.4)
  96. 96
    DeepSeekDossier

    high

    Arena rating

    DeepSeek V4 Flash Preview, Arena rating: 1,431 (scale 1000.0 to 1826.0), 95% confidence interval 1424.1 to 1438.11,431(1424.1–1438.1)
  97. 97
    OpenAIDossier

    Arena rating

    GPT-5.2, Arena rating: 1,419 (scale 1000.0 to 1826.0), 95% confidence interval 1396.2 to 1441.71,419(1396.2–1441.7)
  98. 98
    OpenAIDossier

    medium

    Arena rating

    GPT-5, Arena rating: 1,417 (scale 1000.0 to 1826.0), 95% confidence interval 1401.8 to 1433.01,417(1401.8–1433.0)
  99. 99
    Thinking MachinesDossier

    Arena rating

    Inkling, Arena rating: 1,414 (scale 1000.0 to 1826.0), 95% confidence interval 1408.2 to 1420.11,414(1408.2–1420.1)
  100. 100
    Thinking MachinesDossier

    Arena rating

    Inkling Small, Arena rating: 1,408 (scale 1000.0 to 1826.0), 95% confidence interval 1398.9 to 1417.21,408(1398.9–1417.2)
  101. 101
    OpenAIDossier

    Codex CLI

    Arena rating

    GPT-5.3 Codex, Arena rating: 1,406 (scale 1000.0 to 1826.0), 95% confidence interval 1392.4 to 1419.81,406(1392.4–1419.8)
  102. 102
    Moonshot AIDossier

    instant

    Arena rating

    Kimi K2.5, Arena rating: 1,406 (scale 1000.0 to 1826.0), 95% confidence interval 1393.5 to 1417.71,406(1393.5–1417.7)
  103. 103
    NVIDIADossier

    Arena rating

    Nemotron 3.5 Super VL 120B A12B (NVFP4), Arena rating: 1,401 (scale 1000.0 to 1826.0), 95% confidence interval 1390.5 to 1411.51,401(1390.5–1411.5)
  104. 104
    AlibabaDossier

    Arena rating

    Qwen3.5 397B A17B, Arena rating: 1,400 (scale 1000.0 to 1826.0), 95% confidence interval 1395.0 to 1405.31,400(1395.0–1405.3)
  105. 105
    OpenAIDossier

    Arena rating

    GPT-5.4, Arena rating: 1,400 (scale 1000.0 to 1826.0), 95% confidence interval 1390.0 to 1409.71,400(1390.0–1409.7)
  106. 106
    MiniMaxDossier

    Arena rating

    MiniMax M2.7, Arena rating: 1,398 (scale 1000.0 to 1826.0), 95% confidence interval 1392.1 to 1403.81,398(1392.1–1403.8)
  107. 107
    OpenAIDossier

    high

    Arena rating

    GPT-5.4 mini, Arena rating: 1,398 (scale 1000.0 to 1826.0), 95% confidence interval 1391.2 to 1404.31,398(1391.2–1404.3)
  108. 108
    OpenAIDossier

    medium

    Arena rating

    GPT-5.1, Arena rating: 1,398 (scale 1000.0 to 1826.0), 95% confidence interval 1386.0 to 1409.21,398(1386.0–1409.2)
  109. 109
    AnthropicDossier

    high

    Arena rating

    Sonnet 4.5, Arena rating: 1,394 (scale 1000.0 to 1826.0), 95% confidence interval 1386.1 to 1401.71,394(1386.1–1401.7)
  110. 110
    AnthropicDossier

    Arena rating

    Opus 4.1, Arena rating: 1,390 (scale 1000.0 to 1826.0), 95% confidence interval 1379.3 to 1400.91,390(1379.3–1400.9)
  111. 111
    MiniMaxDossier

    Arena rating

    MiniMax M2.5, Arena rating: 1,386 (scale 1000.0 to 1826.0), 95% confidence interval 1377.4 to 1394.21,386(1377.4–1394.2)
  112. 112
    AnthropicDossier

    Arena rating

    Sonnet 4.5, Arena rating: 1,385 (scale 1000.0 to 1826.0), 95% confidence interval 1377.5 to 1392.21,385(1377.5–1392.2)
  113. 113
    GoogleDossier

    minimal

    Arena rating

    Gemini 3 Flash Preview, Arena rating: 1,385 (scale 1000.0 to 1826.0), 95% confidence interval 1379.3 to 1389.81,385(1379.3–1389.8)
  114. 114
    MiniMaxDossier

    Arena rating

    MiniMax M2.1 Preview, Arena rating: 1,384 (scale 1000.0 to 1826.0), 95% confidence interval 1374.6 to 1393.71,384(1374.6–1393.7)
  115. 115

    thinking

    Arena rating

    Grok 4.20 0309, Arena rating: 1,376 (scale 1000.0 to 1826.0), 95% confidence interval 1369.6 to 1381.81,376(1369.6–1381.8)
  116. 116
    OpenAIDossier

    Codex CLI

    Arena rating

    GPT-5.3 Codex, Arena rating: 1,371 (scale 1000.0 to 1826.0), 95% confidence interval 1359.7 to 1382.41,371(1359.7–1382.4)
  117. 117
    GoogleDossier

    Arena rating

    Gemma 4 31B, Arena rating: 1,370 (scale 1000.0 to 1826.0), 95% confidence interval 1364.3 to 1376.31,370(1364.3–1376.3)
  118. 118
    UpstageDossier

    Arena rating

    Solar Pro 4, Arena rating: 1,369 (scale 1000.0 to 1826.0), 95% confidence interval 1358.5 to 1379.51,369(1358.5–1379.5)
  119. 119
    DeepSeekDossier

    thinking

    Arena rating

    DeepSeek V3.2, Arena rating: 1,362 (scale 1000.0 to 1826.0), 95% confidence interval 1353.3 to 1371.61,362(1353.3–1371.6)
  120. 120
    AlibabaDossier

    Arena rating

    Qwen3.5 122B A10B, Arena rating: 1,360 (scale 1000.0 to 1826.0), 95% confidence interval 1352.5 to 1367.91,360(1352.5–1367.9)
  121. 121
    GoogleDossier

    Arena rating

    Gemma 4 26B A4B, Arena rating: 1,358 (scale 1000.0 to 1826.0), 95% confidence interval 1341.1 to 1375.11,358(1341.1–1375.1)
  122. 122
    TencentDossier

    Arena rating

    Hy3 Preview, Arena rating: 1,358 (scale 1000.0 to 1826.0), 95% confidence interval 1340.4 to 1375.51,358(1340.4–1375.5)
  123. 123

    Arena rating

    Grok 4.3, Arena rating: 1,357 (scale 1000.0 to 1826.0), 95% confidence interval 1351.1 to 1363.71,357(1351.1–1363.7)
  124. 124
    AlibabaDossier

    Arena rating

    Qwen3.5 27B, Arena rating: 1,357 (scale 1000.0 to 1826.0), 95% confidence interval 1349.3 to 1365.51,357(1349.3–1365.5)
  125. 125

    Arena rating

    Muse Glimmer, Arena rating: 1,354 (scale 1000.0 to 1826.0), 95% confidence interval 1338.0 to 1369.51,354(1338.0–1369.5)
  126. 126
    PoolsideDossier

    Arena rating

    Laguna M.1, Arena rating: 1,350 (scale 1000.0 to 1826.0), 95% confidence interval 1340.5 to 1359.21,350(1340.5–1359.2)
  127. 127
    OpenAIDossier

    Arena rating

    GPT-5.1, Arena rating: 1,342 (scale 1000.0 to 1826.0), 95% confidence interval 1333.7 to 1350.81,342(1333.7–1350.8)
  128. 128

    Arena rating

    GLM-4.6, Arena rating: 1,342 (scale 1000.0 to 1826.0), 95% confidence interval 1331.0 to 1352.71,342(1331.0–1352.7)
  129. 129
    OpenAIDossier

    Arena rating

    GPT-5.2 Codex, Arena rating: 1,339 (scale 1000.0 to 1826.0), 95% confidence interval 1329.6 to 1347.61,339(1329.6–1347.6)
  130. 130
    OpenAIDossier

    Arena rating

    GPT-5.1 Codex, Arena rating: 1,338 (scale 1000.0 to 1826.0), 95% confidence interval 1326.5 to 1349.81,338(1326.5–1349.8)
  131. 131
    XiaomiDossier

    non-thinking

    Arena rating

    MiMo-V2-Flash, Arena rating: 1,331 (scale 1000.0 to 1826.0), 95% confidence interval 1321.4 to 1341.01,331(1321.4–1341.0)
  132. 132
    AnthropicDossier

    Arena rating

    Haiku 4.5, Arena rating: 1,330 (scale 1000.0 to 1826.0), 95% confidence interval 1325.2 to 1335.11,330(1325.2–1335.1)
  133. 133
    DeepSeekDossier

    Arena rating

    DeepSeek V3.2, Arena rating: 1,326 (scale 1000.0 to 1826.0), 95% confidence interval 1317.6 to 1333.71,326(1317.6–1333.7)
  134. 134
    Moonshot AIDossier

    Arena rating

    Kimi K2 Thinking Turbo, Arena rating: 1,322 (scale 1000.0 to 1826.0), 95% confidence interval 1314.2 to 1329.31,322(1314.2–1329.3)
  135. 135
    PoolsideDossier

    Arena rating

    Laguna XS.2, Arena rating: 1,304 (scale 1000.0 to 1826.0), 95% confidence interval 1292.8 to 1314.31,304(1292.8–1314.3)
  136. 136
    MiniMaxDossier

    Arena rating

    MiniMax M2, Arena rating: 1,299 (scale 1000.0 to 1826.0), 95% confidence interval 1287.8 to 1309.91,299(1287.8–1309.9)
  137. 137
    XiaomiDossier

    thinking

    Arena rating

    MiMo-V2-Flash, Arena rating: 1,292 (scale 1000.0 to 1826.0), 95% confidence interval 1275.2 to 1308.21,292(1275.2–1308.2)
  138. 138
    AlibabaDossier

    Arena rating

    Qwen3 Coder 480B A35B Instruct, Arena rating: 1,274 (scale 1000.0 to 1826.0), 95% confidence interval 1266.4 to 1282.21,274(1266.4–1282.2)
  139. 139
    DeepSeekDossier

    Arena rating

    DeepSeek V3.2 Exp, Arena rating: 1,272 (scale 1000.0 to 1826.0), 95% confidence interval 1258.2 to 1285.41,272(1258.2–1285.4)
  140. 140
    MistralDossier

    Arena rating

    Mistral Medium 3.5, Arena rating: 1,263 (scale 1000.0 to 1826.0), 95% confidence interval 1248.3 to 1278.11,263(1248.3–1278.1)
  141. 141
    KuaishouDossier

    Arena rating

    KAT-Coder-Pro V1, Arena rating: 1,257 (scale 1000.0 to 1826.0), 95% confidence interval 1237.6 to 1276.71,257(1237.6–1276.7)
  142. 142
    GoogleDossier

    Arena rating

    Gemini 3.1 Flash-Lite Preview, Arena rating: 1,256 (scale 1000.0 to 1826.0), 95% confidence interval 1248.9 to 1262.71,256(1248.9–1262.7)
  143. 143
    AlibabaDossier

    Arena rating

    Qwen3.5 35B A3B, Arena rating: 1,253 (scale 1000.0 to 1826.0), 95% confidence interval 1235.5 to 1270.51,253(1235.5–1270.5)
  144. 144
    AlibabaDossier

    Arena rating

    Qwen3.5 Flash, Arena rating: 1,244 (scale 1000.0 to 1826.0), 95% confidence interval 1224.0 to 1263.41,244(1224.0–1263.4)
  145. 145
    OpenAIDossier

    Arena rating

    GPT-5.1 Codex mini, Arena rating: 1,243 (scale 1000.0 to 1826.0), 95% confidence interval 1221.2 to 1265.81,243(1221.2–1265.8)
  146. 146

    thinking

    Arena rating

    Grok 4.1 Fast, Arena rating: 1,242 (scale 1000.0 to 1826.0), 95% confidence interval 1230.5 to 1252.81,242(1230.5–1252.8)
  147. 147
    Arcee AIDossier

    Arena rating

    Trinity Large Thinking, Arena rating: 1,237 (scale 1000.0 to 1826.0), 95% confidence interval 1216.7 to 1256.81,237(1216.7–1256.8)
  148. 148
    MistralDossier

    Arena rating

    Mistral Large 3, Arena rating: 1,231 (scale 1000.0 to 1826.0), 95% confidence interval 1205.0 to 1256.81,231(1205.0–1256.8)
  149. 149
    GoogleDossier

    Arena rating

    Gemini 2.5 Pro, Arena rating: 1,226 (scale 1000.0 to 1826.0), 95% confidence interval 1210.4 to 1242.31,226(1210.4–1242.3)
  150. 150

    thinking

    Arena rating

    Grok 4.1, Arena rating: 1,216 (scale 1000.0 to 1826.0), 95% confidence interval 1191.1 to 1241.01,216(1191.1–1241.0)
  151. 151
    MistralDossier

    Arena rating

    Devstral 2, Arena rating: 1,195 (scale 1000.0 to 1826.0), 95% confidence interval 1174.7 to 1216.31,195(1174.7–1216.3)
  152. 152

    Arena rating

    Granite 4.1 8B, Arena rating: 1,193 (scale 1000.0 to 1826.0), 95% confidence interval 1174.1 to 1211.31,193(1174.1–1211.3)
  153. 153
    InceptionDossier

    Arena rating

    Mercury 2, Arena rating: 1,171 (scale 1000.0 to 1826.0), 95% confidence interval 1145.8 to 1195.41,171(1145.8–1195.4)
  154. 154

    Arena rating

    Grok Code Fast 1, Arena rating: 1,165 (scale 1000.0 to 1826.0), 95% confidence interval 1136.9 to 1193.91,165(1136.9–1193.9)
  155. 155

    thinking

    Arena rating

    Grok 4 Fast, Arena rating: 1,159 (scale 1000.0 to 1826.0), 95% confidence interval 1131.2 to 1186.21,159(1131.2–1186.2)
  156. 156
    MistralDossier

    Arena rating

    Devstral Medium (2507), Arena rating: 1,071 (scale 1000.0 to 1826.0), 95% confidence interval 1040.3 to 1102.01,071(1040.3–1102.0)
Unmatched upstream rows excluded from this board (1)

These upstream rows could not be mapped cleanly to the canonical model registry or reasoning effort enum.

Raw nameParsed modelReasonDetail
Qwen4-Max (xHigh)Qwen4-Maxno-aliasrating