Skip to content
MCS · Models · 62

Model Ratings.

The tracked cohort, scored and tiered.

62 models ranked against the frontier intelligence recipe · every axis sourced.

Scores are relative to the tracked cohort — swap the recipe or the field and the ranks move. The weights are an editorial call; the axis values underneath are sourced, cell by cell.

Snapshot · verified 1w ago · 2026-10-05

Recipe

Raw reasoning, knowledge, and agentic coding — price-blind.

The board · ranked · Frontier intelligence

Each cell brightens toward the cohort leader and is ringed when it holds it; dimmed rows are sparse (low coverage) — read their composite with care.

  1. 1
    Claude Opus 5.5
    Anthropic
    50% · mediumS96
    1. AA Index58
    2. GPQA—
    3. HLE61.4%
    4. SWE-bench—
    5. Arena1504
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  2. 2
    Gemini 4 Argon
    Google DeepMind
    50% · mediumS90
    1. AA Index53
    2. GPQA—
    3. HLE57.1%
    4. SWE-bench—
    5. Arena1525
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  3. 3
    Claude Fable 5.1
    Anthropic
    67% · mediumS88
    1. AA Index53
    2. GPQA93.7%
    3. HLE59.1%
    4. SWE-bench—
    5. Arena1498
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  4. 4
    Claude Opus 5
    Anthropic
    83% · highS86
    1. AA Index51
    2. GPQA93.2%
    3. HLE54.9%
    4. SWE-bench96%
    5. Arena1493
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  5. 5
    Claude Fable 5
    Anthropic
    83% · highS86
    1. AA Index50
    2. GPQA92.6%
    3. HLE55.5%
    4. SWE-bench95%
    5. Arena1504
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  6. 6
    GPT-6 Astra
    OpenAI
    67% · mediumA85
    1. AA Index53
    2. GPQA96.1%
    3. HLE54.7%
    4. SWE-bench—
    5. Arena1480
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  7. 7
    Claude Sonnet 5.5
    Anthropic
    50% · mediumA83
    1. AA Index56
    2. GPQA—
    3. HLE55%
    4. SWE-bench—
    5. Arena1471
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  8. 8
    GPT-6.1 Sol
    OpenAI
    50% · mediumA78
    1. AA Index52
    2. GPQA—
    3. HLE52.9%
    4. SWE-bench—
    5. Arena1483
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  9. 9
    Muse Spark 1.3
    Meta
    67% · mediumA78
    1. AA Index48
    2. GPQA93.5%
    3. HLE48.7%
    4. SWE-bench—
    5. Arena1493
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  10. 10
    GPT-5.6 Sol
    OpenAI
    67% · mediumA76
    1. AA Index47
    2. GPQA94.1%
    3. HLE49.5%
    4. SWE-bench—
    5. Arena1484
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  11. 11
    Qwen3.8 Max (0902)
    Alibaba (Qwen)
    50% · mediumA72
    1. AA Index45
    2. GPQA92.8%
    3. HLE43.1%
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: aa_index · Weakest: hle

  12. 12
    Kimi K3
    Moonshot AI
    67% · mediumA71
    1. AA Index44
    2. GPQA93.5%
    3. HLE46.9%
    4. SWE-bench—
    5. Arena1485
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  13. 13
    Gemini 3.8 Flash
    Google DeepMind
    67% · mediumA70
    1. AA Index41
    2. GPQA95.3%
    3. HLE47.8%
    4. SWE-bench—
    5. Arena1493
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  14. 14
    Claude Opus 4.7
    Anthropic
    83% · highB70
    1. AA Index41
    2. GPQA94.2%
    3. HLE46.9%
    4. SWE-bench87.6%
    5. Arena1501
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  15. 15
    GLM-5.3
    Z.AI (Zhipu)
    67% · mediumB70
    1. AA Index45
    2. GPQA91.7%
    3. HLE42.3%
    4. SWE-bench—
    5. Arena1483
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  16. 16
    Claude Opus 4.8
    Anthropic
    83% · highB68
    1. AA Index42
    2. GPQA93.6%
    3. HLE45.7%
    4. SWE-bench88.6%
    5. Arena1483
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  17. 17
    GPT-5.6 Terra
    OpenAI
    50% · mediumB68
    1. AA Index42
    2. GPQA92.5%
    3. HLE42.9%
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  18. 18
    MiMo-V2.6-Pro
    Xiaomi
    50% · mediumB67
    1. AA Index46
    2. GPQA—
    3. HLE49.4%
    4. SWE-bench—
    5. Arena1480
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  19. 19
    Step 5 Preview
    StepFun
    67% · mediumB66
    1. AA Index44
    2. GPQA93.5%
    3. HLE46.5%
    4. SWE-bench—
    5. Arena1454
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  20. 20
    Gemini 3.7 Flash
    Google DeepMind
    50% · mediumB66
    1. AA Index39
    2. GPQA94.5%
    3. HLE—
    4. SWE-bench—
    5. Arena1490
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  21. 21
    Grok 4.6
    xAI
    67% · mediumB66
    1. AA Index44
    2. GPQA94.9%
    3. HLE42.9%
    4. SWE-bench—
    5. Arena1456
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  22. 22
    Qwen3.8-Flash-Next
    Alibaba (Qwen)
    67% · mediumB66
    1. AA Index40
    2. GPQA91.7%
    3. HLE35.9%
    4. SWE-bench—
    5. Arena—
    6. LiveCode91.9%

    Strongest: gpqa · Weakest: hle

  23. 23
    Qwen3.8 Max
    Alibaba (Qwen)
    67% · mediumB65
    1. AA Index40
    2. GPQA92.6%
    3. HLE43.6%
    4. SWE-bench—
    5. Arena1481
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  24. 24
    GPT-6 Sol
    OpenAI
    50% · mediumB64
    1. AA Index48
    2. GPQA—
    3. HLE47.9%
    4. SWE-bench—
    5. Arena1457
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  25. 25
    GPT-5.5
    OpenAI
    83% · highB64
    1. AA Index38
    2. GPQA93.5%
    3. HLE44.3%
    4. SWE-bench88.7%
    5. Arena1482
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  26. 26
    GLM-5.3-Flash
    Z.AI (Zhipu)
    67% · mediumB64
    1. AA Index42
    2. GPQA91.2%
    3. HLE39.9%
    4. SWE-bench—
    5. Arena1475
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  27. 27
    GPT-5.6 Luna
    OpenAI
    33% · lowB60
    1. AA Index37
    2. GPQA91.1%
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  28. 28
    DeepSeek V4.1 Flash
    DeepSeek
    67% · mediumB60
    1. AA Index39
    2. GPQA90.9%
    3. HLE39.3%
    4. SWE-bench—
    5. Arena1474
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  29. 29
    DeepSeek V4 Pro 0813
    DeepSeek
    50% · mediumB60
    1. AA Index36
    2. GPQA92.8%
    3. HLE41%
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  30. 30
    Grok 4.7
    xAI
    50% · mediumB56
    1. AA Index46
    2. GPQA—
    3. HLE43.1%
    4. SWE-bench—
    5. Arena1442
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  31. 31
    DeepSeek V4 Flash 0731
    DeepSeek
    50% · mediumC54
    1. AA Index34
    2. GPQA91%
    3. HLE37%
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  32. 32
    DeepSeek V4 Pro
    DeepSeek
    83% · highC52
    1. AA Index30
    2. GPQA90.1%
    3. HLE37.7%
    4. SWE-bench80.6%
    5. Arena—
    6. LiveCode93.5%

    Strongest: gpqa · Weakest: aa_index

  33. 33
    Gemini 3.1 Pro
    Google DeepMind
    83% · highC52
    1. AA Index30
    2. GPQA94.1%
    3. HLE44.7%
    4. SWE-bench80.6%
    5. Arena1487
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  34. 34
    Qwen3.8-27B
    Alibaba (Qwen)
    83% · highC50
    1. AA Index34
    2. GPQA90.5%
    3. HLE33.9%
    4. SWE-bench—
    5. Arena1438
    6. LiveCode90.3%

    Strongest: gpqa · Weakest: arena_elo

  35. 35
    MiniMax-M3
    MiniMax
    50% · mediumC49
    1. AA Index29
    2. GPQA92.9%
    3. HLE39%
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  36. 36
    Grok 4.5
    xAI
    17% · lowC47
    1. AA Index39
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  37. 37
    Gemini 3.5 Flash
    Google DeepMind
    50% · mediumC43
    1. AA Index33
    2. GPQA—
    3. HLE40.2%
    4. SWE-bench—
    5. Arena1478
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  38. 38
    gpt-oss-120b
    OpenAI
    33% · lowC42
    1. AA Index—
    2. GPQA80.1%
    3. HLE14.9%
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: hle

  39. 39
    GPT-6 Luna
    OpenAI
    50% · mediumC42
    1. AA Index38
    2. GPQA—
    3. HLE38.5%
    4. SWE-bench—
    5. Arena1443
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  40. 40
    DeepSeek V4 Flash (Preview)
    DeepSeek
    83% · highC42
    1. AA Index24
    2. GPQA88.1%
    3. HLE34.8%
    4. SWE-bench79%
    5. Arena—
    6. LiveCode91.6%

    Strongest: gpqa · Weakest: aa_index

  41. 41
    Qwen3.7 Max
    Alibaba (Qwen)
    83% · highC41
    1. AA Index29
    2. GPQA92.4%
    3. HLE38.2%
    4. SWE-bench—
    5. Arena1460
    6. LiveCode78.2%

    Strongest: gpqa · Weakest: livecodebench

  42. 42
    MiMo-V2.6-Flash
    Xiaomi
    33% · lowC40
    1. AA Index38
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena1452
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  43. 43
    Gemini 3.6 Flash
    Google DeepMind
    33% · lowD40
    1. AA Index34
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena1480
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  44. 44
    Kimi K2.6
    Moonshot AI
    100% · highD40
    1. AA Index27
    2. GPQA90.5%
    3. HLE34.7%
    4. SWE-bench80.2%
    5. Arena1422
    6. LiveCode89.6%

    Strongest: gpqa · Weakest: arena_elo

  45. 45
    GLM-5.1
    Z.AI (Zhipu)
    67% · mediumD39
    1. AA Index26
    2. GPQA86.2%
    3. HLE31%
    4. SWE-bench—
    5. Arena1467
    6. LiveCode—

    Strongest: gpqa · Weakest: aa_index

  46. 46
    Claude Sonnet 5
    Anthropic
    50% · mediumD38
    1. AA Index38
    2. GPQA—
    3. HLE43.2%
    4. SWE-bench72.7%
    5. Arena—
    6. LiveCode—

    Strongest: aa_index · Weakest: swebench

  47. 47
    Muse Spark
    Meta
    33% · lowD36
    1. AA Index31
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena1488
    6. LiveCode—

    Strongest: arena_elo · Weakest: aa_index

  48. 48
    DeepSeek V4 Flash Vision Exp
    DeepSeek
    17% · lowD36
    1. AA Index35
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  49. 49
    GLM-5.2
    Z.AI (Zhipu)
    17% · lowD33
    1. AA Index34
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  50. 50
    GPT-5.3 Codex
    OpenAI
    17% · lowD31
    1. AA Index33
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  51. 51
    gpt-oss-20b
    OpenAI
    33% · lowD30
    1. AA Index—
    2. GPQA71.5%
    3. HLE10.9%
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: hle

  52. 52
    Phi-4
    Microsoft
    17% · lowD27
    1. AA Index—
    2. GPQA56.1%
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: gpqa

  53. 53
    Claude Sonnet 4.6
    Anthropic
    50% · mediumD24
    1. AA Index30
    2. GPQA—
    3. HLE—
    4. SWE-bench80.2%
    5. Arena1442
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  54. 54
    Qwen3.7 Plus
    Alibaba (Qwen)
    33% · lowD15
    1. AA Index25
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena1456
    6. LiveCode—

    Strongest: arena_elo · Weakest: aa_index

  55. 55
    Kimi K2.7 Code
    Moonshot AI
    17% · lowD11
    1. AA Index26
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  56. 56
    Gemini 3.5 Flash-Lite
    Google DeepMind
    33% · lowD9
    1. AA Index22
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena1456
    6. LiveCode—

    Strongest: arena_elo · Weakest: aa_index

  57. 57
    MiMo-V2.5-Pro
    Xiaomi
    33% · lowD9
    1. AA Index26
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena1425
    6. LiveCode—

    Strongest: aa_index · Weakest: arena_elo

  58. 58
    Mistral Small 3.2 24B
    Mistral AI
    17% · lowD9
    1. AA Index—
    2. GPQA46.1%
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: gpqa

  59. 59
    Grok 4.3
    xAI
    17% · lowD8
    1. AA Index25
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  60. 60
    GPT-5.4 mini
    OpenAI
    17% · lowD6
    1. AA Index24
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: aa_index · Weakest: aa_index

  61. 61
    Gemma 3 27B
    Google
    17% · lowD0
    1. AA Index—
    2. GPQA41.4%
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: gpqa · Weakest: gpqa

  62. 62
    Qwen3-32B
    Alibaba (Qwen)
    0% · lowD0
    1. AA Index—
    2. GPQA—
    3. HLE—
    4. SWE-bench—
    5. Arena—
    6. LiveCode—

    Strongest: — · Weakest: —