Benchmark Almanac

Artificial Analysis Intelligence Index v5

Claude Fable 5.2 64.6 leads since November 2026
Claude Fable 5.2 64.6 leads since November 2026
In use
Version 5.0

Artificial Analysis’s leaderboard hasn’t scored a model released after Gemini 4 Neon (November 2026). Labs report their own numbers for one newer model: Inkling 2 Mini at 20.1.

Version 5 of Artificial Analysis's index: nine evaluations in four equal categories, most with private test sets, run three times each in its own agent harness.

Released 27 October 2026. Knowledge Work: AA-Briefcase v1.1 15%, AutomationBench-AA 5%, GDP.pdf 5%. Coding: FrontierCode 2.0 15%, Terminal-Bench 4.0 10%. Science: Terminal-Bench-Science 15%, CritPt v2 hard subset 10%. General: AA-Omniscience v1.1 15% (accuracy 10%, non-hallucination 5%), HLE-Diamond 10%. Dropped GDPval-AA, SciCode, AA-LCR and Humanity's Last Exam. Earlier versions have their own records. Scores do not compare across versions. Its numbers here come from Artificial Analysis's API, with its permission, along with the scores labs quote in their own posts.

Released
October 2026
Kind
Composite indices
Scored
Checked automatically
Answers
Agent task
Measures
Capability
Language
English
Interaction
Agentic
Environment
Terminal, APIs and tools
Items from
Another benchmark
Run by
A host
Strongest model tested
GPT-6.2 Astra #1 on the capability index
Last in a launch post
November 2026
Last model added
November 2026
Scores refreshed
November 19th, 2026
84
Entries on Artificial Analysis’s leaderboard, 69 models we track
7
Launch posts list it, from 3 labs
3
System cards report it, from 2 labs

Claude Fable 5.2 leads with 64.6 Claude Fable 5.2 leads with 64.6

Best score per entry from Artificial Analysis’s leaderboard, read on November 19th, 2026. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.

Claude Fable 5.2 leads with 64.6, claims included

Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model. Striped bars are claims, and dotted bars are reruns by Epoch AI or another board.

84 entries. Scroll sideways for the rest.

  1. 64.6 Claude Fable 5.2
  2. 62.1 GPT-6.2 Astra
  3. 61.4 Gemini 4 Neon
  4. 60.0 Gemini 4 Carbon
  5. 59.1 Claude Opus 5.5
  6. 57.3 GPT-6.2 Sol
  7. 56.9 GPT-6 Astra
  8. 54.6 GPT-6.1 Sol
  9. 54.5 Claude Sonnet 5.5
  10. 53.9 Gemini 4 Hydrogen
  11. 51.3 GPT-6.1 Terra
  12. 50.4 Gemini 4 Argon
  13. 50.1 Muse Wonder Full
  14. 50.0 Claude Fable 5.1
  15. 46.5 Kimi K3.1 Pro
  16. 44.2 Claude Opus 5
  17. 44.2 GPT-6 Sol
  18. 43.6 Muse Wonder
  19. 42.9 Qwen4-Max
  20. 36.6 Muse Spark 1.3
  21. 36.5 Claude Haiku 5.5
  22. 36.0 Muse Ember
  23. 35.5 Grok 4.7
  24. 34.6 Tencent Hy4
  25. 33.7 Qwen3.8 Max (0902)
  26. 32.6 DeepSeek V4.1 Flash 1017
  27. 32.3 GPT-5.6 Terra
  28. 31.7 GLM-5.3
  29. 31.0 Grok 4.6
  30. 30.8 Gemini 3.8 Flash
  31. 30.6 Inkling 2 Small
  32. 29.9 MiMo-V2.6 Pro
  33. 29.6 Ling 3.1 Flash 1019
  34. 29.1 MiniMax-M3.1-Flash
  35. 28.6 Qwen3.8-Flash-Next
  36. 28.5 Step 5 Preview
  37. 28.0 JoyAI-VL-Preview
  38. 27.9 GPT-6 Luna
  39. 27.8 GLM-5.3-Flash
  40. 26.7 MiMo-V2.6 Flash
  41. 26.1 Mistral Large 4
  42. 25.3 Mistral Large 4 Preview
  43. 24.9 Muse Spark Open
  44. 24.0 Agnes 2.5 Pro Theta
  45. 23.8 GPT-5.6 Luna
  46. 23.1 DeepSeek V4 Pro 0813
  47. 20.4 Nanbeige4.3-33B
  48. 20.1 Qwen 3.8 27B
  49. 19.0 JT-4.1 Flash 236B A21B
  50. 18.9 MiniMax-M3
  51. 17.8 K2 Horizon 375B A23B
  52. 17.7 Beam 1.0
  53. 17.4 Kimi K3
  54. 14.2 Solar Pro 4
  55. 13.9 K2 Horizon 32B
  56. 13.8 K2 Horizon MoVA 36B A4B
  57. 13.7 Apodex 1.1
  58. 13.6 Nemotron 3.5 Super VL
  59. 12.9 Inkling-Small
  60. 12.5 Inkling
  61. 11.8 Nemotron 3 Ultra 550B A55B (Reasoning)
  62. 11.7 Gemini 3.5 Flash-Lite
  63. 10.6 K2 Horizon 7B
  64. 9.0 A.X-K2
  65. 8.0 K2 Horizon 3.7B
  66. 7.6 K-EXAONE 2.0
  67. 7.4 Claude Haiku 4.5
  68. 6.3 Muse Glimmer
  69. 6.2 Gemini 3.1 Flash-Lite
  70. 6.2 Command A+
  71. 6.0 LFM3-1.7B
  72. 5.5 Nemotron 3.5 Lightning 30B-A3B
  73. 5.5 Granite 4.2 30B
  74. 5.1 Granite 4.2 8B
  75. 4.8 Gemma 4 E4B
  76. 4.7 Granite 4.2 3B
  77. 4.6 Mistral Medium 3.5
  78. 4.4 Gemma 4 31B IT
  79. 4.3 Mistral Small 4
  80. 3.8 gpt-oss-120b
  81. 3.4 North Mini Code 1.0
  82. 2.7 K2 Horizon 0.9B
  83. 1.8 Nemotron 3 Nano 4B
  84. 1.7 LFM3-VL-2B
  1. 1 Claude Fable 5.2 max, default fallback 64.6
  2. 2 GPT-6.2 Astra max 62.1
  3. 3 Gemini 4 Neon xhigh 61.4
  4. 4 Gemini 4 Carbon high 60.0
  5. 5 Claude Opus 5.5 max, default fallback 59.1
  6. 6 GPT-6.2 Sol max 57.3
  7. 7 GPT-6 Astra max 56.9
  8. 8 GPT-6.1 Sol max 54.6
  9. 9 Claude Sonnet 5.5 max, default fallback 54.5
  10. 10 Gemini 4 Hydrogen xhigh 53.9
  11. 11 GPT-6.1 Terra max 51.3
  12. 12 Gemini 4 Argon high 50.4
  13. 13 Muse Wonder Full max 50.1
  14. 14 Claude Fable 5.1 xhigh, default fallback 50.0
  15. 15 Kimi K3.1 Pro 46.5
  16. 16 Claude Opus 5 max 44.2
  17. 17 GPT-6 Sol max 44.2
  18. 18 Muse Wonder max 43.6
  19. 19 Qwen4-Max 42.9
  20. 20 Muse Spark 1.3 max 36.6
  21. 21 Claude Haiku 5.5 max 36.5
  22. 22 Muse Ember max 36.0
  23. 23 Grok 4.7 xhigh 35.5
  24. 24 Tencent Hy4 34.6
  25. 25 Qwen3.8 Max (0902) 33.7
  26. 26 DeepSeek V4.1 Flash 1017 max 32.6
  27. 27 GPT-5.6 Terra max 32.3
  28. 28 GLM-5.3 max 31.7
  29. 29 Grok 4.6 xhigh 31.0
  30. 30 Gemini 3.8 Flash high 30.8
  31. 31 Inkling 2 Small 30.6
  32. 32 MiMo-V2.6 Pro 29.9
  33. 33 Ling 3.1 Flash 1019 29.6
  34. 34 MiniMax-M3.1-Flash 29.1
  35. 35 Qwen3.8-Flash-Next 28.6
  36. 36 Step 5 Preview 28.5
  37. 37 JoyAI-VL-Preview 28.0
  38. 38 GPT-6 Luna max 27.9
  39. 39 GLM-5.3-Flash 27.8
  40. 40 MiMo-V2.6 Flash 26.7
  41. 41 Mistral Large 4 high 26.1
  42. 42 Mistral Large 4 Preview 25.3
  43. 43 Muse Spark Open xhigh 24.9
  44. 44 Agnes 2.5 Pro Theta 24.0
  45. 45 GPT-5.6 Luna max 23.8
  46. 46 DeepSeek V4 Pro 0813 max 23.1
  47. 47 Nanbeige4.3-33B 20.4
  48. 48 Qwen 3.8 27B xhigh 20.1
  49. 49 JT-4.1 Flash 236B A21B 19.0
  50. 50 MiniMax-M3 18.9
  51. 51 K2 Horizon 375B A23B 17.8
  52. 52 Beam 1.0 17.7
  53. 53 Kimi K3 low 17.4
  54. 54 Solar Pro 4 14.2
  55. 55 K2 Horizon 32B 13.9
  56. 56 K2 Horizon MoVA 36B A4B 13.8
  57. 57 Apodex 1.1 13.7
  58. 58 Nemotron 3.5 Super VL 13.6
  59. 59 Inkling-Small 12.9
  60. 60 Inkling xhigh 12.5
  61. 61 Nemotron 3 Ultra 550B A55B (Reasoning) 11.8
  62. 62 Gemini 3.5 Flash-Lite 11.7
  63. 63 K2 Horizon 7B 10.6
  64. 64 A.X-K2 9.0
  65. 65 K2 Horizon 3.7B 8.0
  66. 66 K-EXAONE 2.0 7.6
  67. 67 Claude Haiku 4.5 reasoning 7.4
  68. 68 Muse Glimmer high 6.3
  69. 69 Gemini 3.1 Flash-Lite 6.2
  70. 70 Command A+ 6.2
  71. 71 LFM3-1.7B 6.0
  72. 72 Nemotron 3.5 Lightning 30B-A3B 5.5
  73. 73 Granite 4.2 30B 5.5
  74. 74 Granite 4.2 8B 5.1
  75. 75 Gemma 4 E4B reasoning 4.8
  76. 76 Granite 4.2 3B 4.7
  77. 77 Mistral Medium 3.5 4.6
  78. 78 Gemma 4 31B IT reasoning 4.4
  79. 79 Mistral Small 4 reasoning 4.3
  80. 80 gpt-oss-120b high 3.8
  81. 81 North Mini Code 1.0 3.4
  82. 82 K2 Horizon 0.9B 2.7
  83. 83 Nemotron 3 Nano 4B 1.8
  84. 84 LFM3-VL-2B 1.7

Source: Artificial Analysis’s leaderboard

  1. 1 Claude Fable 5.2 max, default fallback 64.6
  2. 2 GPT-6.2 Astra max 62.1
  3. 3 Gemini 4 Neon xhigh 61.4
  4. 4 Gemini 4 Carbon high 60.0
  5. 5 Claude Opus 5.5 max, default fallback 59.1
  6. 6 GPT-6.2 Sol max 57.3
  7. 7 GPT-6 Astra max 56.9
  8. 8 GPT-6.1 Sol max 54.6
  9. 9 Claude Sonnet 5.5 max, default fallback 54.5
  10. 10 Gemini 4 Hydrogen xhigh 53.9
  11. 11 GPT-6.1 Terra max 51.3
  12. 12 Gemini 4 Argon high 50.4
  13. 13 Muse Wonder Full max 50.1
  14. 14 Claude Fable 5.1 xhigh, default fallback 50.0
  15. 15 Kimi K3.1 Pro 46.5
  16. 16 Claude Opus 5 max 44.2
  17. 17 GPT-6 Sol max 44.2
  18. 18 Muse Wonder max 43.6
  19. 19 Qwen4-Max 42.9
  20. 20 Muse Spark 1.3 max 36.6
  21. 21 Claude Haiku 5.5 max 36.5
  22. 22 Muse Ember max 36.0
  23. 23 Grok 4.7 xhigh 35.5
  24. 24 Tencent Hy4 34.6
  25. 25 Qwen3.8 Max (0902) 33.7
  26. 26 DeepSeek V4.1 Flash 1017 max 32.6
  27. 27 GPT-5.6 Terra max 32.3
  28. 28 GLM-5.3 max 31.7
  29. 29 Grok 4.6 xhigh 31.0
  30. 30 Gemini 3.8 Flash high 30.8
  31. 31 Inkling 2 Small 30.6
  32. 32 MiMo-V2.6 Pro 29.9
  33. 33 MiniMax-M3.1-Flash 29.1
  34. 34 Qwen3.8-Flash-Next 28.6
  35. 35 Step 5 Preview 28.5
  36. 36 GPT-6 Luna max 27.9
  37. 37 GLM-5.3-Flash 27.8
  38. 38 MiMo-V2.6 Flash 26.7
  39. 39 Mistral Large 4 high 26.1
  40. 40 Mistral Large 4 Preview 25.3
  41. 41 Muse Spark Open xhigh 24.9
  42. 42 GPT-5.6 Luna max 23.8
  43. 43 DeepSeek V4 Pro 0813 max 23.1
  44. 44 Qwen 3.8 27B xhigh 20.1
  45. 45 MiniMax-M3 18.9
  46. 46 Beam 1.0 17.7
  47. 47 Kimi K3 low 17.4
  48. 48 Solar Pro 4 14.2
  49. 49 Nemotron 3.5 Super VL 13.6
  50. 50 Inkling-Small 12.9
  51. 51 Inkling xhigh 12.5
  52. 52 Gemini 3.5 Flash-Lite 11.7
  53. 53 K-EXAONE 2.0 7.6
  54. 54 Claude Haiku 4.5 reasoning 7.4
  55. 55 Muse Glimmer high 6.3
  56. 56 Gemini 3.1 Flash-Lite 6.2
  57. 57 Command A+ 6.2
  58. 58 LFM3-1.7B 6.0
  59. 59 Nemotron 3.5 Lightning 30B-A3B 5.5
  60. 60 Granite 4.2 30B 5.5
  61. 61 Granite 4.2 8B 5.1
  62. 62 Gemma 4 E4B reasoning 4.8
  63. 63 Granite 4.2 3B 4.7
  64. 64 Mistral Medium 3.5 4.6
  65. 65 Gemma 4 31B IT reasoning 4.4
  66. 66 Mistral Small 4 reasoning 4.3
  67. 67 gpt-oss-120b high 3.8
  68. 68 North Mini Code 1.0 3.4
  69. 69 Nemotron 3 Nano 4B 1.8

Source: Artificial Analysis’s leaderboard

  1. 1 Claude Fable 5.2 max, default fallback 64.6
  2. 2 GPT-6.2 Astra max 62.1
  3. 3 Gemini 4 Neon xhigh 61.4
  4. 4 Gemini 4 Carbon high 60.0
  5. 5 Claude Opus 5.5 max, default fallback 59.1
  6. 6 GPT-6.2 Sol max 57.3
  7. 7 GPT-6 Astra max 56.9
  8. 8 GPT-6.1 Sol max 54.6
  9. 9 Claude Sonnet 5.5 max, default fallback 54.5
  10. 10 Gemini 4 Hydrogen xhigh effort 54.0
  11. 11 GPT-6.1 Terra max 51.3
  12. 12 Gemini 4 Argon high 50.4
  13. 13 Muse Wonder Full max 50.1
  14. 14 Claude Fable 5.1 xhigh, default fallback 50.0
  15. 15 Kimi K3.1 Pro 46.5
  16. 16 Claude Opus 5 max 44.2
  17. 17 GPT-6 Sol max 44.2
  18. 18 Muse Wonder max 43.6
  19. 19 Qwen4-Max 42.9
  20. 20 Muse Spark 1.3 max 36.6
  21. 21 Claude Haiku 5.5 max 36.5
  22. 22 Muse Ember max 36.0
  23. 23 Grok 4.7 xhigh 35.5
  24. 24 Tencent Hy4 34.6
  25. 25 Qwen3.8 Max (0902) 33.7
  26. 26 DeepSeek V4.1 Flash 1017 max 32.6
  27. 27 GPT-5.6 Terra max 32.3
  28. 28 GLM-5.3 max 31.7
  29. 29 Grok 4.6 xhigh 31.0
  30. 30 Gemini 3.8 Flash high 30.8
  31. 31 Inkling 2 Small 30.6
  32. 32 MiMo-V2.6 Pro 29.9
  33. 33 MiniMax-M3.1-Flash 29.1
  34. 34 Qwen3.8-Flash-Next 28.6
  35. 35 Step 5 Preview 28.5
  36. 36 GPT-6 Luna max 27.9
  37. 37 GLM-5.3-Flash 27.8
  38. 38 MiMo-V2.6 Flash 26.7
  39. 39 Mistral Large 4 high 26.1
  40. 40 Mistral Large 4 Preview 25.3
  41. 41 Muse Spark Open xhigh 24.9
  42. 42 GPT-5.6 Luna max 23.8
  43. 43 DeepSeek V4 Pro 0813 max 23.1
  44. 44 Qwen 3.8 27B xhigh 20.1
  45. 45 Inkling 2 Mini 20.1
  46. 46 MiniMax-M3 18.9
  47. 47 Beam 1.0 17.7
  48. 48 Kimi K3 low 17.4
  49. 49 Solar Pro 4 14.2
  50. 50 Nemotron 3.5 Super VL 13.6
  51. 51 Inkling-Small 12.9
  52. 52 Inkling xhigh 12.5
  53. 53 Gemini 3.5 Flash-Lite 11.7
  54. 54 K-EXAONE 2.0 7.6
  55. 55 Claude Haiku 4.5 reasoning 7.4
  56. 56 Muse Glimmer high 6.3
  57. 57 Gemini 3.1 Flash-Lite 6.2
  58. 58 Command A+ 6.2
  59. 59 LFM3-1.7B 6.0
  60. 60 Nemotron 3.5 Lightning 30B-A3B 5.5
  61. 61 Granite 4.2 30B 5.5
  62. 62 Granite 4.2 8B 5.1
  63. 63 Gemma 4 E4B reasoning 4.8
  64. 64 Granite 4.2 3B 4.7
  65. 65 Mistral Medium 3.5 4.6
  66. 66 Gemma 4 31B IT reasoning 4.4
  67. 67 Mistral Small 4 reasoning 4.3
  68. 68 gpt-oss-120b high 3.8
  69. 69 North Mini Code 1.0 3.4
  70. 70 Nemotron 3 Nano 4B 1.8

Sources: Artificial Analysis’s leaderboard and labs’ model cards

#EntryScoreSettingDateSource
1 Claude Fable 5.2 64.6 max, default fallback Nov 2026 Official leaderboard
2 GPT-6.2 Astra 62.1 max Nov 2026 Official leaderboard
3 Gemini 4 Neon 61.4 xhigh Nov 2026 Official leaderboard
4 Gemini 4 Carbon 60.0 high Oct 2026 Official leaderboard
5 Claude Opus 5.5 59.1 max, default fallback Sep 2026 Official leaderboard
6 GPT-6.2 Sol 57.3 max Nov 2026 Official leaderboard
7 GPT-6 Astra 56.9 max Sep 2026 Official leaderboard
8 GPT-6.1 Sol 54.6 max Sep 2026 Official leaderboard
9 Claude Sonnet 5.5 54.5 max, default fallback Sep 2026 Official leaderboard
10 Gemini 4 Hydrogen 53.9 xhigh Nov 2026 Official leaderboard
11 GPT-6.1 Terra 51.3 max Oct 2026 Official leaderboard
12 Gemini 4 Argon 50.4 high Sep 2026 Official leaderboard
13 Muse Wonder Full 50.1 max Oct 2026 Official leaderboard
14 Claude Fable 5.1 50.0 xhigh, default fallback Sep 2026 Official leaderboard
15 Kimi K3.1 Pro 46.5 Oct 2026 Official leaderboard
16 Claude Opus 5 44.2 max Jul 2026 Official leaderboard
17 GPT-6 Sol 44.2 max Sep 2026 Official leaderboard
18 Muse Wonder 43.6 max Oct 2026 Official leaderboard
19 Qwen4-Max 42.9 Nov 2026 Official leaderboard
20 Muse Spark 1.3 36.6 max Sep 2026 Official leaderboard
21 Claude Haiku 5.5 36.5 max Oct 2026 Official leaderboard
22 Muse Ember 36.0 max Nov 2026 Official leaderboard
23 Grok 4.7 35.5 xhigh Sep 2026 Official leaderboard
24 Tencent Hy4 34.6 Nov 2026 Official leaderboard
25 Qwen3.8 Max (0902) 33.7 Sep 2026 Official leaderboard
26 DeepSeek V4.1 Flash 1017 32.6 max Oct 2026 Official leaderboard
27 GPT-5.6 Terra 32.3 max Jul 2026 Official leaderboard
28 GLM-5.3 31.7 max Aug 2026 Official leaderboard
29 Grok 4.6 31.0 xhigh Aug 2026 Official leaderboard
30 Gemini 3.8 Flash 30.8 high Sep 2026 Official leaderboard
31 Inkling 2 Small 30.6 Nov 2026 Official leaderboard
32 MiMo-V2.6 Pro 29.9 Sep 2026 Official leaderboard
33 Ling 3.1 Flash 1019 29.6 Official leaderboard
34 MiniMax-M3.1-Flash 29.1 Oct 2026 Official leaderboard
35 Qwen3.8-Flash-Next 28.6 Aug 2026 Official leaderboard
36 Step 5 Preview 28.5 Sep 2026 Official leaderboard
37 JoyAI-VL-Preview 28.0 Official leaderboard
38 GPT-6 Luna 27.9 max Sep 2026 Official leaderboard
39 GLM-5.3-Flash 27.8 Aug 2026 Official leaderboard
40 MiMo-V2.6 Flash 26.7 Sep 2026 Official leaderboard
41 Mistral Large 4 26.1 high Oct 2026 Official leaderboard
42 Mistral Large 4 Preview 25.3 Oct 2026 Official leaderboard
43 Muse Spark Open 24.9 xhigh Oct 2026 Official leaderboard
44 Agnes 2.5 Pro Theta 24.0 Official leaderboard
45 GPT-5.6 Luna 23.8 max Jul 2026 Official leaderboard
46 DeepSeek V4 Pro 0813 23.1 max Aug 2026 Official leaderboard
47 Nanbeige4.3-33B 20.4 Official leaderboard
48 Qwen 3.8 27B 20.1 xhigh Aug 2026 Official leaderboard
49 JT-4.1 Flash 236B A21B 19.0 Official leaderboard
50 MiniMax-M3 18.9 Jun 2026 Official leaderboard
51 K2 Horizon 375B A23B 17.8 Official leaderboard
52 Beam 1.0 17.7 Oct 2026 Official leaderboard
53 Kimi K3 17.4 low Jul 2026 Official leaderboard
54 Solar Pro 4 14.2 Aug 2026 Official leaderboard
55 K2 Horizon 32B 13.9 Official leaderboard
56 K2 Horizon MoVA 36B A4B 13.8 Official leaderboard
57 Apodex 1.1 13.7 Official leaderboard
58 Nemotron 3.5 Super VL 13.6 Oct 2026 Official leaderboard
59 Inkling-Small 12.9 Jul 2026 Official leaderboard
60 Inkling 12.5 xhigh Jul 2026 Official leaderboard
61 Nemotron 3 Ultra 550B A55B (Reasoning) 11.8 Official leaderboard
62 Gemini 3.5 Flash-Lite 11.7 Jul 2026 Official leaderboard
63 K2 Horizon 7B 10.6 Official leaderboard
64 A.X-K2 9.0 Official leaderboard
65 K2 Horizon 3.7B 8.0 Official leaderboard
66 K-EXAONE 2.0 7.6 Jul 2026 Official leaderboard
67 Claude Haiku 4.5 7.4 reasoning Oct 2025 Official leaderboard
68 Muse Glimmer 6.3 high Aug 2026 Official leaderboard
69 Gemini 3.1 Flash-Lite 6.2 Mar 2026 Official leaderboard
70 Command A+ 6.2 May 2026 Official leaderboard
71 LFM3-1.7B 6.0 Oct 2026 Official leaderboard
72 Nemotron 3.5 Lightning 30B-A3B 5.5 Aug 2026 Official leaderboard
73 Granite 4.2 30B 5.5 Aug 2026 Official leaderboard
74 Granite 4.2 8B 5.1 Aug 2026 Official leaderboard
75 Gemma 4 E4B 4.8 reasoning Mar 2026 Official leaderboard
76 Granite 4.2 3B 4.7 Aug 2026 Official leaderboard
77 Mistral Medium 3.5 4.6 Apr 2026 Official leaderboard
78 Gemma 4 31B IT 4.4 reasoning Apr 2026 Official leaderboard
79 Mistral Small 4 4.3 reasoning Mar 2026 Official leaderboard
80 gpt-oss-120b 3.8 high Aug 2025 Official leaderboard
81 North Mini Code 1.0 3.4 Jun 2026 Official leaderboard
82 K2 Horizon 0.9B 2.7 Official leaderboard
83 Nemotron 3 Nano 4B 1.8 Mar 2026 Official leaderboard
84 LFM3-VL-2B 1.7 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Fable 5.2 64.6 max, default fallback Nov 2026 Official leaderboard
2 GPT-6.2 Astra 62.1 max Nov 2026 Official leaderboard
3 Gemini 4 Neon 61.4 xhigh Nov 2026 Official leaderboard
4 Gemini 4 Carbon 60.0 high Oct 2026 Official leaderboard
5 Claude Opus 5.5 59.1 max, default fallback Sep 2026 Official leaderboard
6 GPT-6.2 Sol 57.3 max Nov 2026 Official leaderboard
7 GPT-6 Astra 56.9 max Sep 2026 Official leaderboard
8 GPT-6.1 Sol 54.6 max Sep 2026 Official leaderboard
9 Claude Sonnet 5.5 54.5 max, default fallback Sep 2026 Official leaderboard
10 Gemini 4 Hydrogen 53.9 xhigh Nov 2026 Official leaderboard
11 GPT-6.1 Terra 51.3 max Oct 2026 Official leaderboard
12 Gemini 4 Argon 50.4 high Sep 2026 Official leaderboard
13 Muse Wonder Full 50.1 max Oct 2026 Official leaderboard
14 Claude Fable 5.1 50.0 xhigh, default fallback Sep 2026 Official leaderboard
15 Kimi K3.1 Pro 46.5 Oct 2026 Official leaderboard
16 Claude Opus 5 44.2 max Jul 2026 Official leaderboard
17 GPT-6 Sol 44.2 max Sep 2026 Official leaderboard
18 Muse Wonder 43.6 max Oct 2026 Official leaderboard
19 Qwen4-Max 42.9 Nov 2026 Official leaderboard
20 Muse Spark 1.3 36.6 max Sep 2026 Official leaderboard
21 Claude Haiku 5.5 36.5 max Oct 2026 Official leaderboard
22 Muse Ember 36.0 max Nov 2026 Official leaderboard
23 Grok 4.7 35.5 xhigh Sep 2026 Official leaderboard
24 Tencent Hy4 34.6 Nov 2026 Official leaderboard
25 Qwen3.8 Max (0902) 33.7 Sep 2026 Official leaderboard
26 DeepSeek V4.1 Flash 1017 32.6 max Oct 2026 Official leaderboard
27 GPT-5.6 Terra 32.3 max Jul 2026 Official leaderboard
28 GLM-5.3 31.7 max Aug 2026 Official leaderboard
29 Grok 4.6 31.0 xhigh Aug 2026 Official leaderboard
30 Gemini 3.8 Flash 30.8 high Sep 2026 Official leaderboard
31 Inkling 2 Small 30.6 Nov 2026 Official leaderboard
32 MiMo-V2.6 Pro 29.9 Sep 2026 Official leaderboard
33 MiniMax-M3.1-Flash 29.1 Oct 2026 Official leaderboard
34 Qwen3.8-Flash-Next 28.6 Aug 2026 Official leaderboard
35 Step 5 Preview 28.5 Sep 2026 Official leaderboard
36 GPT-6 Luna 27.9 max Sep 2026 Official leaderboard
37 GLM-5.3-Flash 27.8 Aug 2026 Official leaderboard
38 MiMo-V2.6 Flash 26.7 Sep 2026 Official leaderboard
39 Mistral Large 4 26.1 high Oct 2026 Official leaderboard
40 Mistral Large 4 Preview 25.3 Oct 2026 Official leaderboard
41 Muse Spark Open 24.9 xhigh Oct 2026 Official leaderboard
42 GPT-5.6 Luna 23.8 max Jul 2026 Official leaderboard
43 DeepSeek V4 Pro 0813 23.1 max Aug 2026 Official leaderboard
44 Qwen 3.8 27B 20.1 xhigh Aug 2026 Official leaderboard
45 MiniMax-M3 18.9 Jun 2026 Official leaderboard
46 Beam 1.0 17.7 Oct 2026 Official leaderboard
47 Kimi K3 17.4 low Jul 2026 Official leaderboard
48 Solar Pro 4 14.2 Aug 2026 Official leaderboard
49 Nemotron 3.5 Super VL 13.6 Oct 2026 Official leaderboard
50 Inkling-Small 12.9 Jul 2026 Official leaderboard
51 Inkling 12.5 xhigh Jul 2026 Official leaderboard
52 Gemini 3.5 Flash-Lite 11.7 Jul 2026 Official leaderboard
53 K-EXAONE 2.0 7.6 Jul 2026 Official leaderboard
54 Claude Haiku 4.5 7.4 reasoning Oct 2025 Official leaderboard
55 Muse Glimmer 6.3 high Aug 2026 Official leaderboard
56 Gemini 3.1 Flash-Lite 6.2 Mar 2026 Official leaderboard
57 Command A+ 6.2 May 2026 Official leaderboard
58 LFM3-1.7B 6.0 Oct 2026 Official leaderboard
59 Nemotron 3.5 Lightning 30B-A3B 5.5 Aug 2026 Official leaderboard
60 Granite 4.2 30B 5.5 Aug 2026 Official leaderboard
61 Granite 4.2 8B 5.1 Aug 2026 Official leaderboard
62 Gemma 4 E4B 4.8 reasoning Mar 2026 Official leaderboard
63 Granite 4.2 3B 4.7 Aug 2026 Official leaderboard
64 Mistral Medium 3.5 4.6 Apr 2026 Official leaderboard
65 Gemma 4 31B IT 4.4 reasoning Apr 2026 Official leaderboard
66 Mistral Small 4 4.3 reasoning Mar 2026 Official leaderboard
67 gpt-oss-120b 3.8 high Aug 2025 Official leaderboard
68 North Mini Code 1.0 3.4 Jun 2026 Official leaderboard
69 Nemotron 3 Nano 4B 1.8 Mar 2026 Official leaderboard
#ModelScoreSettingDateSource
1 Claude Fable 5.2 64.6 max, default fallback Nov 2026 Official leaderboard
2 GPT-6.2 Astra 62.1 max Nov 2026 Official leaderboard
3 Gemini 4 Neon 61.4 xhigh Nov 2026 Official leaderboard
4 Gemini 4 Carbon 60.0 high Oct 2026 Official leaderboard
5 Claude Opus 5.5 59.1 max, default fallback Sep 2026 Official leaderboard
6 GPT-6.2 Sol 57.3 max Nov 2026 Official leaderboard
7 GPT-6 Astra 56.9 max Sep 2026 Official leaderboard
8 GPT-6.1 Sol 54.6 max Sep 2026 Official leaderboard
9 Claude Sonnet 5.5 54.5 max, default fallback Sep 2026 Official leaderboard
10 Gemini 4 Hydrogen 54.0 xhigh effort Nov 2026 Lab model card
11 GPT-6.1 Terra 51.3 max Oct 2026 Official leaderboard
12 Gemini 4 Argon 50.4 high Sep 2026 Official leaderboard
13 Muse Wonder Full 50.1 max Oct 2026 Official leaderboard
14 Claude Fable 5.1 50.0 xhigh, default fallback Sep 2026 Official leaderboard
15 Kimi K3.1 Pro 46.5 Oct 2026 Official leaderboard
16 Claude Opus 5 44.2 max Jul 2026 Official leaderboard
17 GPT-6 Sol 44.2 max Sep 2026 Official leaderboard
18 Muse Wonder 43.6 max Oct 2026 Official leaderboard
19 Qwen4-Max 42.9 Nov 2026 Official leaderboard
20 Muse Spark 1.3 36.6 max Sep 2026 Official leaderboard
21 Claude Haiku 5.5 36.5 max Oct 2026 Official leaderboard
22 Muse Ember 36.0 max Nov 2026 Official leaderboard
23 Grok 4.7 35.5 xhigh Sep 2026 Official leaderboard
24 Tencent Hy4 34.6 Nov 2026 Official leaderboard
25 Qwen3.8 Max (0902) 33.7 Sep 2026 Official leaderboard
26 DeepSeek V4.1 Flash 1017 32.6 max Oct 2026 Official leaderboard
27 GPT-5.6 Terra 32.3 max Jul 2026 Official leaderboard
28 GLM-5.3 31.7 max Aug 2026 Official leaderboard
29 Grok 4.6 31.0 xhigh Aug 2026 Official leaderboard
30 Gemini 3.8 Flash 30.8 high Sep 2026 Official leaderboard
31 Inkling 2 Small 30.6 Nov 2026 Official leaderboard
32 MiMo-V2.6 Pro 29.9 Sep 2026 Official leaderboard
33 MiniMax-M3.1-Flash 29.1 Oct 2026 Official leaderboard
34 Qwen3.8-Flash-Next 28.6 Aug 2026 Official leaderboard
35 Step 5 Preview 28.5 Sep 2026 Official leaderboard
36 GPT-6 Luna 27.9 max Sep 2026 Official leaderboard
37 GLM-5.3-Flash 27.8 Aug 2026 Official leaderboard
38 MiMo-V2.6 Flash 26.7 Sep 2026 Official leaderboard
39 Mistral Large 4 26.1 high Oct 2026 Official leaderboard
40 Mistral Large 4 Preview 25.3 Oct 2026 Official leaderboard
41 Muse Spark Open 24.9 xhigh Oct 2026 Official leaderboard
42 GPT-5.6 Luna 23.8 max Jul 2026 Official leaderboard
43 DeepSeek V4 Pro 0813 23.1 max Aug 2026 Official leaderboard
44 Qwen 3.8 27B 20.1 xhigh Aug 2026 Official leaderboard
45 Inkling 2 Mini 20.1 Nov 2026 Lab model card
46 MiniMax-M3 18.9 Jun 2026 Official leaderboard
47 Beam 1.0 17.7 Oct 2026 Official leaderboard
48 Kimi K3 17.4 low Jul 2026 Official leaderboard
49 Solar Pro 4 14.2 Aug 2026 Official leaderboard
50 Nemotron 3.5 Super VL 13.6 Oct 2026 Official leaderboard
51 Inkling-Small 12.9 Jul 2026 Official leaderboard
52 Inkling 12.5 xhigh Jul 2026 Official leaderboard
53 Gemini 3.5 Flash-Lite 11.7 Jul 2026 Official leaderboard
54 K-EXAONE 2.0 7.6 Jul 2026 Official leaderboard
55 Claude Haiku 4.5 7.4 reasoning Oct 2025 Official leaderboard
56 Muse Glimmer 6.3 high Aug 2026 Official leaderboard
57 Gemini 3.1 Flash-Lite 6.2 Mar 2026 Official leaderboard
58 Command A+ 6.2 May 2026 Official leaderboard
59 LFM3-1.7B 6.0 Oct 2026 Official leaderboard
60 Nemotron 3.5 Lightning 30B-A3B 5.5 Aug 2026 Official leaderboard
61 Granite 4.2 30B 5.5 Aug 2026 Official leaderboard
62 Granite 4.2 8B 5.1 Aug 2026 Official leaderboard
63 Gemma 4 E4B 4.8 reasoning Mar 2026 Official leaderboard
64 Granite 4.2 3B 4.7 Aug 2026 Official leaderboard
65 Mistral Medium 3.5 4.6 Apr 2026 Official leaderboard
66 Gemma 4 31B IT 4.4 reasoning Apr 2026 Official leaderboard
67 Mistral Small 4 4.3 reasoning Mar 2026 Official leaderboard
68 gpt-oss-120b 3.8 high Aug 2025 Official leaderboard
69 North Mini Code 1.0 3.4 Jun 2026 Official leaderboard
70 Nemotron 3 Nano 4B 1.8 Mar 2026 Official leaderboard

Labs report scores for 7 models in their cards. Switch the scores above to rank them together.

Rankings by part

We read 6 parts from Artificial Analysis’s leaderboard, each ranking models on one part of the test, such as Finance and accounting, Strategy and operations and Legal. Claude Fable 5.2 leads 4 of them, and Gemini 4 Neon leads the other 2. Arrows show how far a model moved from its overall rank among the models we track.

Finance and accounting

Artificial Analysis's finance and accounting index, built from the Intelligence Index evaluations that test this kind of work. 69 models we track have a score here.

  1. 1 Gemini 4 Neon ↑2 65.1
  2. 2 Claude Fable 5.2 ↓1 64.7
  3. 3 Gemini 4 Carbon ↑1 62.6
  4. 4 GPT-6.2 Astra ↓2 60.9
  5. 5 Claude Opus 5.5 59.4
  6. 6 Gemini 4 Hydrogen ↑4 58.1
  7. 7 GPT-6 Astra 57.4
  8. 8 GPT-6.2 Sol ↓2 57.2
  9. 9 GPT-6.1 Sol ↓1 56.2
  10. 10 Gemini 4 Argon ↑2 55.9

More effort adds a median 11 index

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Claude Fable 5.2
low, defaul… high, defau…
64.6
GPT-6.2 Astra
low medium xhigh
62.1
Gemini 4 Neon
low medium high
61.4
Gemini 4 Carbon
low medium
60.0
Claude Opus 5.5
low, defaul… medium, def… high, defau… xhigh, defa…
59.1
GPT-6.2 Sol
low medium max
57.3
GPT-6 Astra
low medium xhigh
56.9
GPT-6.1 Sol
low medium high max
54.6
Claude Sonnet 5.5
low, defaul… medium, def… high, defau… xhigh, defa… max, defaul…
54.5
Gemini 4 Hydrogen
low medium high
53.9
GPT-6.1 Terra
non-reasoni… medium high max
51.3
Gemini 4 Argon
medium
50.4

Source: Artificial Analysis’s leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Fable 5.2 max, default fallback 64.6
Claude Fable 5.2 xhigh, default fallback 62.2
GPT-6.2 Astra max 62.1
Claude Fable 5.2 high, default fallback 61.8
GPT-6.2 Astra xhigh 61.6
Gemini 4 Neon xhigh 61.4
GPT-6.2 Astra high 60.1
Gemini 4 Carbon high 60.0
Gemini 4 Neon high 59.6
Claude Opus 5.5 max, default fallback 59.1
Gemini 4 Carbon medium 58.3
Claude Opus 5.5 xhigh, default fallback 57.4
GPT-6.2 Sol max 57.3
GPT-6.2 Astra medium 57.0
GPT-6 Astra max 56.9
Claude Fable 5.2 medium, default fallback 56.4
GPT-6.2 Sol xhigh 55.7
GPT-6 Astra xhigh 55.3
GPT-6.2 Sol high 54.9
Gemini 4 Neon medium 54.9
GPT-6 Astra high 54.8
GPT-6.1 Sol max 54.6
Claude Sonnet 5.5 max, default fallback 54.5
Gemini 4 Carbon low 54.2
Claude Fable 5.2 low, default fallback 53.9
Gemini 4 Hydrogen xhigh 53.9
GPT-6.2 Sol medium 52.7
GPT-6.1 Sol xhigh 52.5
Claude Opus 5.5 high, default fallback 51.8
GPT-6 Astra medium 51.6
GPT-6.1 Terra max 51.3
Gemini 4 Hydrogen high 50.8
Gemini 4 Argon high 50.4
GPT-6.1 Sol high 50.3
Muse Wonder Full max 50.1
Claude Fable 5.1 xhigh, default fallback 50.0
Claude Sonnet 5.5 xhigh, default fallback 49.9
Claude Fable 5.1 max, default fallback 49.2
Claude Opus 5.5 medium, default fallback 48.2
Gemini 4 Argon medium 47.6
Claude Fable 5.1 high, default fallback 47.2
Kimi K3.1 Pro 46.5
GPT-6.1 Sol medium 46.3
GPT-6.2 Astra low 45.7
GPT-6.1 Terra xhigh 45.3
Claude Opus 5 max 44.2
GPT-6 Sol max 44.2
Gemini 4 Neon low 44.1
Muse Wonder max 43.6
GPT-6 Astra low 43.3
Claude Fable 5.1 medium, default fallback 43.2
Qwen4-Max 42.9
GPT-6.1 Terra high 42.4
Claude Sonnet 5.5 high, default fallback 41.5
Gemini 4 Hydrogen medium 41.2
GPT-6.2 Sol low 40.6
Muse Wonder high 40.5
Claude Fable 5.1 low, default fallback 38.7
GPT-6.1 Terra medium 38.2
Muse Spark 1.3 max 36.6
Claude Haiku 5.5 max 36.5
Muse Ember max 36.0
Grok 4.7 xhigh 35.5
Grok 4.7 high 35.2
Claude Opus 5.5 low, default fallback 34.8
GPT-6.1 Sol low 34.8
Claude Haiku 5.5 xhigh 34.6
Tencent Hy4 34.6
Muse Ember xhigh 34.1
Qwen3.8 Max (0902) 33.7
DeepSeek V4.1 Flash 1017 max 32.6
GPT-5.6 Terra max 32.3
GLM-5.3 max 31.7
Grok 4.7 low 31.1
Grok 4.6 xhigh 31.0
Gemini 3.8 Flash high 30.8
Claude Sonnet 5.5 medium, default fallback 30.8
Inkling 2 Small 30.6
GPT-6.1 Terra non-reasoning 30.5
Gemini 3.8 Flash medium 30.1
MiMo-V2.6 Pro 29.9
Claude Haiku 5.5 high 29.9
GPT-6.1 Terra low 29.4
MiniMax-M3.1-Flash 29.1
Qwen3.8-Flash-Next 28.6
DeepSeek V4.1 Flash 1017 high 28.6
Step 5 Preview 28.5
GPT-6 Luna max 27.9
GLM-5.3-Flash 27.8
Gemini 4 Hydrogen low 26.9
MiMo-V2.6 Flash 26.7
Mistral Large 4 high 26.1
Claude Sonnet 5.5 low, default fallback 25.6
Mistral Large 4 Preview 25.3
Muse Spark Open xhigh 24.9
Claude Haiku 5.5 medium 24.6
GPT-5.6 Luna max 23.8
GPT-6 Luna xhigh 23.5
DeepSeek V4 Pro 0813 max 23.1
GLM-5.3 low 23.1
Mistral Large 4 medium 23.1
Gemini 3.8 Flash low 22.5
GPT-6 Luna high 21.4
DeepSeek V4 Pro 0813 high 20.8
Qwen 3.8 27B xhigh 20.1
MiniMax-M3 18.9
Claude Haiku 5.5 low 18.6
GPT-6 Luna medium 18.5
DeepSeek V4.1 Flash 1017 non-reasoning 18.5
Beam 1.0 17.7
Qwen 3.8 27B medium 17.5
Kimi K3 low 17.4
Qwen 3.8 27B low 15.7
Solar Pro 4 14.2
Nemotron 3.5 Super VL 13.6
Inkling-Small 12.9
DeepSeek V4 Pro 0813 non-reasoning 12.6
Inkling xhigh 12.5
Gemini 3.5 Flash-Lite 11.7
Inkling high 11.0
Qwen 3.8 27B non-reasoning 10.0
GPT-6 Luna low 9.7
Inkling-Small high 9.6
GPT-6 Luna non-reasoning 9.4
K-EXAONE 2.0 7.6
Claude Haiku 4.5 reasoning 7.4
Muse Glimmer high 6.3
Gemini 3.1 Flash-Lite 6.2
Command A+ 6.2
LFM3-1.7B 6.0
Nemotron 3.5 Lightning 30B-A3B 5.5
Granite 4.2 30B 5.5
Granite 4.2 8B 5.1
Gemma 4 E4B reasoning 4.8
Granite 4.2 3B 4.7
Mistral Medium 3.5 4.6
Gemma 4 31B IT reasoning 4.4
Mistral Small 4 reasoning 4.3
gpt-oss-120b high 3.8
North Mini Code 1.0 3.4
Nemotron 3 Nano 4B 1.8

More effort adds a median 11 index

Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.

Lowest effort to highest, light to dark
Claude Fable 5.2
low, defaul… high, defau…
64.6
GPT-6.2 Astra
low medium xhigh
62.1
Gemini 4 Neon
low medium high
61.4
Gemini 4 Carbon
low medium
60.0
Claude Opus 5.5
low, defaul… medium, def… high, defau… xhigh, defa…
59.1
GPT-6.2 Sol
low medium max
57.3
GPT-6 Astra
low medium xhigh
56.9
GPT-6.1 Sol
low medium high max
54.6
Claude Sonnet 5.5
low, defaul… medium, def… high, defau… xhigh, defa… max, defaul…
54.5
Gemini 4 Hydrogen
low medium high
53.9
GPT-6.1 Terra
non-reasoni… medium high max
51.3
Gemini 4 Argon
medium
50.4

Source: Artificial Analysis’s leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Fable 5.2 max, default fallback 64.6
Claude Fable 5.2 xhigh, default fallback 62.2
GPT-6.2 Astra max 62.1
Claude Fable 5.2 high, default fallback 61.8
GPT-6.2 Astra xhigh 61.6
Gemini 4 Neon xhigh 61.4
GPT-6.2 Astra high 60.1
Gemini 4 Carbon high 60.0
Gemini 4 Neon high 59.6
Claude Opus 5.5 max, default fallback 59.1
Gemini 4 Carbon medium 58.3
Claude Opus 5.5 xhigh, default fallback 57.4
GPT-6.2 Sol max 57.3
GPT-6.2 Astra medium 57.0
GPT-6 Astra max 56.9
Claude Fable 5.2 medium, default fallback 56.4
GPT-6.2 Sol xhigh 55.7
GPT-6 Astra xhigh 55.3
GPT-6.2 Sol high 54.9
Gemini 4 Neon medium 54.9
GPT-6 Astra high 54.8
GPT-6.1 Sol max 54.6
Claude Sonnet 5.5 max, default fallback 54.5
Gemini 4 Carbon low 54.2
Claude Fable 5.2 low, default fallback 53.9
Gemini 4 Hydrogen xhigh 53.9
GPT-6.2 Sol medium 52.7
GPT-6.1 Sol xhigh 52.5
Claude Opus 5.5 high, default fallback 51.8
GPT-6 Astra medium 51.6
GPT-6.1 Terra max 51.3
Gemini 4 Hydrogen high 50.8
Gemini 4 Argon high 50.4
GPT-6.1 Sol high 50.3
Muse Wonder Full max 50.1
Claude Fable 5.1 xhigh, default fallback 50.0
Claude Sonnet 5.5 xhigh, default fallback 49.9
Claude Fable 5.1 max, default fallback 49.2
Claude Opus 5.5 medium, default fallback 48.2
Gemini 4 Argon medium 47.6
Claude Fable 5.1 high, default fallback 47.2
Kimi K3.1 Pro 46.5
GPT-6.1 Sol medium 46.3
GPT-6.2 Astra low 45.7
GPT-6.1 Terra xhigh 45.3
Claude Opus 5 max 44.2
GPT-6 Sol max 44.2
Gemini 4 Neon low 44.1
Muse Wonder max 43.6
GPT-6 Astra low 43.3
Claude Fable 5.1 medium, default fallback 43.2
Qwen4-Max 42.9
GPT-6.1 Terra high 42.4
Claude Sonnet 5.5 high, default fallback 41.5
Gemini 4 Hydrogen medium 41.2
GPT-6.2 Sol low 40.6
Muse Wonder high 40.5
Claude Fable 5.1 low, default fallback 38.7
GPT-6.1 Terra medium 38.2
Muse Spark 1.3 max 36.6
Claude Haiku 5.5 max 36.5
Muse Ember max 36.0
Grok 4.7 xhigh 35.5
Grok 4.7 high 35.2
Claude Opus 5.5 low, default fallback 34.8
GPT-6.1 Sol low 34.8
Claude Haiku 5.5 xhigh 34.6
Tencent Hy4 34.6
Muse Ember xhigh 34.1
Qwen3.8 Max (0902) 33.7
DeepSeek V4.1 Flash 1017 max 32.6
GPT-5.6 Terra max 32.3
GLM-5.3 max 31.7
Grok 4.7 low 31.1
Grok 4.6 xhigh 31.0
Gemini 3.8 Flash high 30.8
Claude Sonnet 5.5 medium, default fallback 30.8
Inkling 2 Small 30.6
GPT-6.1 Terra non-reasoning 30.5
Gemini 3.8 Flash medium 30.1
MiMo-V2.6 Pro 29.9
Claude Haiku 5.5 high 29.9
GPT-6.1 Terra low 29.4
MiniMax-M3.1-Flash 29.1
Qwen3.8-Flash-Next 28.6
DeepSeek V4.1 Flash 1017 high 28.6
Step 5 Preview 28.5
GPT-6 Luna max 27.9
GLM-5.3-Flash 27.8
Gemini 4 Hydrogen low 26.9
MiMo-V2.6 Flash 26.7
Mistral Large 4 high 26.1
Claude Sonnet 5.5 low, default fallback 25.6
Mistral Large 4 Preview 25.3
Muse Spark Open xhigh 24.9
Claude Haiku 5.5 medium 24.6
GPT-5.6 Luna max 23.8
GPT-6 Luna xhigh 23.5
DeepSeek V4 Pro 0813 max 23.1
GLM-5.3 low 23.1
Mistral Large 4 medium 23.1
Gemini 3.8 Flash low 22.5
GPT-6 Luna high 21.4
DeepSeek V4 Pro 0813 high 20.8
Qwen 3.8 27B xhigh 20.1
MiniMax-M3 18.9
Claude Haiku 5.5 low 18.6
GPT-6 Luna medium 18.5
DeepSeek V4.1 Flash 1017 non-reasoning 18.5
Beam 1.0 17.7
Qwen 3.8 27B medium 17.5
Kimi K3 low 17.4
Qwen 3.8 27B low 15.7
Solar Pro 4 14.2
Nemotron 3.5 Super VL 13.6
Inkling-Small 12.9
DeepSeek V4 Pro 0813 non-reasoning 12.6
Inkling xhigh 12.5
Gemini 3.5 Flash-Lite 11.7
Inkling high 11.0
Qwen 3.8 27B non-reasoning 10.0
GPT-6 Luna low 9.7
Inkling-Small high 9.6
GPT-6 Luna non-reasoning 9.4
K-EXAONE 2.0 7.6
Claude Haiku 4.5 reasoning 7.4
Muse Glimmer high 6.3
Gemini 3.1 Flash-Lite 6.2
Command A+ 6.2
LFM3-1.7B 6.0
Nemotron 3.5 Lightning 30B-A3B 5.5
Granite 4.2 30B 5.5
Granite 4.2 8B 5.1
Gemma 4 E4B reasoning 4.8
Granite 4.2 3B 4.7
Mistral Medium 3.5 4.6
Gemma 4 31B IT reasoning 4.4
Mistral Small 4 reasoning 4.3
gpt-oss-120b high 3.8
North Mini Code 1.0 3.4
Nemotron 3 Nano 4B 1.8

3 models have held number one on Artificial Analysis’s leaderboard since its release in October 2026 3 models have held number one with claims included since its release in October 2026

Lab claims count from the day the card came out, independent runs from the model’s release. We draw one row each for Artificial Analysis’s leaderboard and what labs claim in their cards.

Anthropic Google DeepMind OpenAI
Artificial Analysis leaderboard 3 leaders
Lab claims 2 leaders

Sources: Artificial Analysis’s leaderboard and labs’ model cards

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On Artificial Analysis’s leaderboard Reported by the lab in its model card
0 20 40 60 80 2026 2027 0 20 40 60 80 2026 2027

Sources: Artificial Analysis’s leaderboard and labs’ model cards

3 labs put it in a launch post in the last six months

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

Google DeepMind
Thinking Machines
OpenAI
DateLaunch postScoreAs the post writes it
Nov 20, 2026 Inkling 2 Mini 20.1 AA Index (v5.0)
Nov 18, 2026 Gemini 4 Neon 61 AA Index v5.0.1
Nov 17, 2026 GPT-6.2 Sol 57.3 Artificial Analysis Intelligence Index v5.0.1 (GPT-6.2 Sol Max), cited from Artificial Analysis
Nov 17, 2026 GPT-6.2 Astra 62.1 Artificial Analysis Intelligence Index v5.0.1 (GPT-6.2 Astra Max), cited from Artificial Analysis
Nov 12, 2026 Gemini 4 Hydrogen 54 Artificial Analysis Intelligence Index, 54 at the xhigh thinking level (51 at the default high level)
Nov 5, 2026 Inkling 2 Small 30.6 AA Index (v5.0)
Oct 28, 2026 Gemini 4 Carbon 59 AA Index v5.0

3 system cards report it, from 2 labs

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Nov 20, 2026 Inkling 2 Mini model card 20.1 AA Index (v5.0) Capabilities
Nov 18, 2026 Gemini 4 Neon model card 61 AA Index v5.0.1 (results table, p.4) Capabilities
Nov 10, 2026 Gemini 4 Carbon model card 59 AA Index v5.0 (results table) Capabilities

6 lab claims next to the leaderboard

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. Every benchmark.

Model Lab says Board says Gap
Gemini 4 Carbon 59.0 AA Index v5.0 60.0 Leaderboard, high −1.0
Gemini 4 Neon 61.0 AA Index v5.0.1 61.4 Leaderboard, xhigh −0.4
Gemini 4 Hydrogen 54.0 Artificial Analysis Intelligence Index, 54 at the xhigh thinking level (51 at the default high level) 53.9 Leaderboard, xhigh Same
Inkling 2 Small 30.6 AA Index (v5.0) 30.6 Leaderboard Same
GPT-6.2 Astra 62.1 Artificial Analysis Intelligence Index v5.0.1 (GPT-6.2 Astra Max), cited from Artificial Analysis 62.1 Leaderboard, max Same
GPT-6.2 Sol 57.3 Artificial Analysis Intelligence Index v5.0.1 (GPT-6.2 Sol Max), cited from Artificial Analysis 57.3 Leaderboard, max Same

Versions

More in Other

All 69 in Other
  • Personal Superintelligence Index (Meta) In use Meta's composite of seven personal and knowledge-work evals: Agents' Last Exam, AutomationBench-AA, FocusBench, JobBench, GDP.pdf, VBT and DRACO.
  • Vals Science Index In use Agents are scored on the equal average of BioMysteryBench, MysteryMechanism and Terminal-Bench-Science. GPT-6.2 Astra 68.4%
  • Decision Index Public v0.3 In use Decision models return probabilities over labeled options in public tasks covering knowledge, language, retrieval, tools, and arts. A chance-corrected weighted aggregate combines the task scores. Granite Switch 4.1 30B 51.3
  • SAGE v1.1 (Vals) In use Models grade handwritten undergraduate math and computer science work against rubrics. Scored by class-balanced accuracy of each rubric check against expert grading. Claude Opus 5.5 91.3%
  • MTEB Fading Massive Text Embedding Benchmark: retrieval, clustering, classification across languages and modalities. Qwen3-Embedding-8B 72.2%
  • SPEED-Bench Only in system cards Measures speculative decoding across varied prompts and domains. Reports the average acceptance length of draft tokens, so higher means faster generation at the same output. Qwen3-Next-80B-A3B-Instruct 2.81

Data audit

10 of 11 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark

Catalog record 1 warning 1 pass
  • Warning Task access We don’t know whether the tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (84 entries), launch posts (7 scores) and system cards (3 scores).
  • Pass Models scored 70 models we track, plus 15 other entries on its leaderboard.
  • Pass Newest score The newest is from November 2026, for Inkling 2 Mini.
  • Pass Frontier models 10 of the top 10 have a score, led by GPT-6.2 Astra (#1).
  • Note Headroom Scored as an index, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on Artificial Analysis: 141 rows for models we track and 15 other entries at the last read.
  • Pass Last read November 19th, 2026.
  • Pass Read worked The last read worked.
Consistency 2 pass
  • Pass Lab claims All 6 pairs agree within 10 points.
  • Pass Known issues No published audit found problems.

How we check