Benchmark Almanac

Cyber ECI

Claude Opus 5.5 171.1 leads since September 2026
Not in model cards

Epoch estimates model capability from cybersecurity benchmark scores and places the result on the general index scale.

Epoch AI publishes this domain index and documents the separate fit for model capability.

Made by
Epoch AI
Kind
Composite indices
Level
Expert
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Another benchmark
Refresh
One fixed set
Run by
Anyone
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
October 2026
Scores refreshed
November 21st, 2026
75
Entries on its own leaderboard, 65 models we track

Claude Opus 5.5 leads with 171.1 Claude Opus 5.5 leads with 171.1

Best score per entry from the benchmark's own leaderboard, read on November 21st, 2026. Where a model ran with several settings or harnesses, we keep its best. 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.

75 entries. Scroll sideways for the rest.

  1. 171.1 Claude Opus 5.5
  2. 171.0 GPT-6 Astra
  3. 169.6 Claude Mythos 5.1
  4. 168.9 GPT-6.1 Sol
  5. 168.3 Claude Mythos 5
  6. 167.5 Claude Opus 5
  7. 167.3 Claude Sonnet 5.5
  8. 166.5 GPT-5.6 Sol
  9. 164.8 Claude Mythos Preview (April)
  10. 164.2 GPT-6 Sol
  11. 163.8 GPT-5.5-Cyber
  12. 162.2 GPT-5.5
  13. 162.2 Claude Mythos Preview (Early)
  14. 160.1 GLM-5.3
  15. 158.9 Claude Haiku 5.5
  16. 158.8 Claude Opus 4.8
  17. 158.0 Claude Opus 4.6
  18. 157.6 Kimi K3
  19. 157.3 GPT-6 Luna
  20. 157.0 GPT-5.4
  21. 156.8 GPT-5.3 Codex
  22. 156.4 GPT-5.4-Cyber
  23. 155.5 Claude Sonnet 5
  24. 154.9 Claude Opus 4.7
  25. 154.2 Claude Sonnet 4.6
  26. 154.2 GPT-5.2 Codex
  27. 154.1 GLM-5.2
  28. 153.7 GPT-5.1-Codex-Max
  29. 153.6 Gemini 3.1 Pro
  30. 152.9 GLM-5.1
  31. 152.9 Claude Opus 4.5
  32. 152.6 GPT-5.2
  33. 152.6 Kimi K2.6
  34. 150.8 Claude Haiku 4.5
  35. 150.3 DeepSeek-V4-Pro
  36. 150.2 GPT-5.4 Mini
  37. 149.4 GPT-5
  38. 149.2 Claude Sonnet 4.5
  39. 148.0 GPT-5.1
  40. 147.9 GLM-5
  41. 145.8 Claude Opus 4.1
  42. 145.8 o3
  43. 145.2 GPT-5 mini
  44. 144.8 Grok 4
  45. 144.7 Claude Opus 4
  46. 144.2 o4-mini
  47. 143.8 Grok 4.1
  48. 143.8 Claude Sonnet 4
  49. 143.1 gpt-oss-120b
  50. 142.7 Kimi K2 Thinking
  51. 141.5 Grok 4 Fast
  52. 141.4 Gemini 2.5 Pro
  53. 140.5 GPT-4.1
  54. 140.2 DeepSeek-V3.1
  55. 140.0 DeepSeek-R1
  56. 139.8 Claude 3.7 Sonnet
  57. 139.7 o1
  58. 139.3 o3-mini
  59. 138.0 Gemini 3 Pro
  60. 137.5 GPT-4.5
  61. 137.3 Claude 3.5 Sonnet (snapshot unspecified)
  62. 137.3 Muse Spark
  63. 136.6 DeepSeek-V3 (snapshot unspecified)
  64. 135.2 GPT-4o
  65. 134.0 DeepSeek-R1 (snapshot unspecified)
  66. 133.8 o1-mini
  67. 133.8 o1-preview
  68. 131.9 Mixtral 8x22B
  69. 131.9 Gemini 1.5 Pro (snapshot unspecified)
  70. 131.9 Llama 3.1-405B
  71. 129.2 Llama 3-70B
  72. 126.7 GPT-4o (snapshot unspecified)
  73. 126.5 Claude 3 Opus
  74. 124.5 GPT-4 Turbo (snapshot unspecified)
  75. 112.2 GPT-3.5 Turbo (snapshot unspecified)
  1. 1 Claude Opus 5.5 171.1
  2. 2 GPT-6 Astra 171.0
  3. 3 Claude Mythos 5.1 169.6
  4. 4 GPT-6.1 Sol 168.9
  5. 5 Claude Mythos 5 168.3
  6. 6 Claude Opus 5 167.5
  7. 7 Claude Sonnet 5.5 167.3
  8. 8 GPT-5.6 Sol 166.5
  9. 9 Claude Mythos Preview (April) 164.8
  10. 10 GPT-6 Sol 164.2
  11. 11 GPT-5.5-Cyber 163.8
  12. 12 GPT-5.5 162.2
  13. 13 Claude Mythos Preview (Early) 162.2
  14. 14 GLM-5.3 160.1
  15. 15 Claude Haiku 5.5 158.9
  16. 16 Claude Opus 4.8 158.8
  17. 17 Claude Opus 4.6 158.0
  18. 18 Kimi K3 157.6
  19. 19 GPT-6 Luna 157.3
  20. 20 GPT-5.4 157.0
  21. 21 GPT-5.3 Codex 156.8
  22. 22 GPT-5.4-Cyber 156.4
  23. 23 Claude Sonnet 5 155.5
  24. 24 Claude Opus 4.7 154.9
  25. 25 Claude Sonnet 4.6 154.2
  26. 26 GPT-5.2 Codex 154.2
  27. 27 GLM-5.2 154.1
  28. 28 GPT-5.1-Codex-Max 153.7
  29. 29 Gemini 3.1 Pro 153.6
  30. 30 GLM-5.1 152.9
  31. 31 Claude Opus 4.5 152.9
  32. 32 GPT-5.2 152.6
  33. 33 Kimi K2.6 152.6
  34. 34 Claude Haiku 4.5 150.8
  35. 35 DeepSeek-V4-Pro 150.3
  36. 36 GPT-5.4 Mini 150.2
  37. 37 GPT-5 149.4
  38. 38 Claude Sonnet 4.5 149.2
  39. 39 GPT-5.1 148.0
  40. 40 GLM-5 147.9
  41. 41 Claude Opus 4.1 145.8
  42. 42 o3 145.8
  43. 43 GPT-5 mini 145.2
  44. 44 Grok 4 144.8
  45. 45 Claude Opus 4 144.7
  46. 46 o4-mini 144.2
  47. 47 Grok 4.1 143.8
  48. 48 Claude Sonnet 4 143.8
  49. 49 gpt-oss-120b 143.1
  50. 50 Kimi K2 Thinking 142.7
  51. 51 Grok 4 Fast 141.5
  52. 52 Gemini 2.5 Pro Jun 2025 141.4
  53. 53 GPT-4.1 140.5
  54. 54 DeepSeek-V3.1 140.2
  55. 55 DeepSeek-R1 May 2025 140.0
  56. 56 Claude 3.7 Sonnet 139.8
  57. 57 o1 139.7
  58. 58 o3-mini 139.3
  59. 59 Gemini 3 Pro 138.0
  60. 60 GPT-4.5 137.5
  61. 61 Claude 3.5 Sonnet (snapshot unspecified) 137.3
  62. 62 Muse Spark 137.3
  63. 63 DeepSeek-V3 (snapshot unspecified) 136.6
  64. 64 GPT-4o Nov 2024 135.2
  65. 65 DeepSeek-R1 (snapshot unspecified) 134.0
  66. 66 o1-mini 133.8
  67. 67 o1-preview 133.8
  68. 68 Mixtral 8x22B 131.9
  69. 69 Gemini 1.5 Pro (snapshot unspecified) 131.9
  70. 70 Llama 3.1-405B 131.9
  71. 71 Llama 3-70B 129.2
  72. 72 GPT-4o (snapshot unspecified) 126.7
  73. 73 Claude 3 Opus 126.5
  74. 74 GPT-4 Turbo (snapshot unspecified) 124.5
  75. 75 GPT-3.5 Turbo (snapshot unspecified) 112.2

Source: Official Cyber ECI leaderboard

Source: Official Cyber ECI leaderboard

#EntryScore ±SettingReleasedRun onSource
1 Claude Opus 5.5 171.1 166.7 to 177.3 Sep 2026 Sep 2026 Official leaderboard
2 GPT-6 Astra 171.0 166.9 to 178 Sep 2026 Sep 2026 Official leaderboard
3 Claude Mythos 5.1 169.6 165.4 to 174.6 Sep 2026 Sep 2026 Official leaderboard
4 GPT-6.1 Sol 168.9 164.9 to 174.8 Sep 2026 Sep 2026 Official leaderboard
5 Claude Mythos 5 168.3 164.2 to 178.5 Jun 2026 Jun 2026 Official leaderboard
6 Claude Opus 5 167.5 163 to 171.6 Jul 2026 Jul 2026 Official leaderboard
7 Claude Sonnet 5.5 167.3 163.3 to 171.7 Sep 2026 Sep 2026 Official leaderboard
8 GPT-5.6 Sol 166.5 163 to 171.3 Jul 2026 Jul 2026 Official leaderboard
9 Claude Mythos Preview (April) 164.8 161.8 to 170 Apr 2026 Official leaderboard
10 GPT-6 Sol 164.2 155 to 167.5 Sep 2026 Sep 2026 Official leaderboard
11 GPT-5.5-Cyber 163.8 160.5 to 169.9 Jun 2026 May 2026 Official leaderboard
12 GPT-5.5 162.2 159.1 to 166.8 Apr 2026 Apr 2026 Official leaderboard
13 Claude Mythos Preview (Early) 162.2 159 to 166.4 Apr 2026 Official leaderboard
14 GLM-5.3 160.1 155.8 to 167 Aug 2026 Aug 2026 Official leaderboard
15 Claude Haiku 5.5 158.9 144.6 to 161.8 Oct 2026 Oct 2026 Official leaderboard
16 Claude Opus 4.8 158.8 155.8 to 163.1 May 2026 May 2026 Official leaderboard
17 Claude Opus 4.6 158.0 155.8 to 160.4 Feb 2026 Feb 2026 Official leaderboard
18 Kimi K3 157.6 148 to 160.7 Jul 2026 Jul 2026 Official leaderboard
19 GPT-6 Luna 157.3 109.2 to 159.9 Sep 2026 Sep 2026 Official leaderboard
20 GPT-5.4 157.0 155 to 159.6 Mar 2026 Mar 2026 Official leaderboard
21 GPT-5.3 Codex 156.8 ±2.4 Feb 2026 Feb 2026 Official leaderboard
22 GPT-5.4-Cyber 156.4 153.9 to 159.4 Apr 2026 Official leaderboard
23 Claude Sonnet 5 155.5 135.3 to 158.4 Jun 2026 Jun 2026 Official leaderboard
24 Claude Opus 4.7 154.9 153 to 157 Apr 2026 Apr 2026 Official leaderboard
25 Claude Sonnet 4.6 154.2 149.9 to 157.5 Feb 2026 Feb 2026 Official leaderboard
26 GPT-5.2 Codex 154.2 151.7 to 155.9 Dec 2025 Dec 2025 Official leaderboard
27 GLM-5.2 154.1 ±2.4 Jun 2026 Jun 2026 Official leaderboard
28 GPT-5.1-Codex-Max 153.7 150 to 155.2 Nov 2025 Nov 2025 Official leaderboard
29 Gemini 3.1 Pro 153.6 150.2 to 156 Feb 2026 Feb 2026 Official leaderboard
30 GLM-5.1 152.9 144.5 to 155.3 Apr 2026 Apr 2026 Official leaderboard
31 Claude Opus 4.5 152.9 150.8 to 155.1 Nov 2025 Nov 2025 Official leaderboard
32 GPT-5.2 152.6 150.2 to 154.8 Dec 2025 Dec 2025 Official leaderboard
33 Kimi K2.6 152.6 148.3 to 155.6 Apr 2026 Apr 2026 Official leaderboard
34 Claude Haiku 4.5 150.8 125.7 to 153.9 Oct 2025 Oct 2025 Official leaderboard
35 DeepSeek-V4-Pro 150.3 148 to 152.4 Apr 2026 Apr 2026 Official leaderboard
36 GPT-5.4 Mini 150.2 145.9 to 153 Mar 2026 Mar 2026 Official leaderboard
37 GPT-5 149.4 147.3 to 150.3 Aug 2025 Aug 2025 Official leaderboard
38 Claude Sonnet 4.5 149.2 147.3 to 150.9 Sep 2025 Sep 2025 Official leaderboard
39 GPT-5.1 148.0 144.9 to 150.3 Nov 2025 Nov 2025 Official leaderboard
40 GLM-5 147.9 143.2 to 151.7 Feb 2026 Feb 2026 Official leaderboard
41 Claude Opus 4.1 145.8 143.2 to 148.6 Aug 2025 Aug 2025 Official leaderboard
42 o3 145.8 142.2 to 147.9 Apr 2025 Apr 2025 Official leaderboard
43 GPT-5 mini 145.2 142.7 to 146.9 Aug 2025 Aug 2025 Official leaderboard
44 Grok 4 144.8 142.6 to 146.2 Jul 2025 Jul 2025 Official leaderboard
45 Claude Opus 4 144.7 142.3 to 146.7 May 2025 May 2025 Official leaderboard
46 o4-mini 144.2 131.7 to 147.1 Apr 2025 Apr 2025 Official leaderboard
47 Grok 4.1 143.8 141.7 to 145.2 Nov 2025 Nov 2025 Official leaderboard
48 Claude Sonnet 4 143.8 141.3 to 147 May 2025 May 2025 Official leaderboard
49 gpt-oss-120b 143.1 139.7 to 145.3 Aug 2025 Aug 2025 Official leaderboard
50 Kimi K2 Thinking 142.7 139 to 145.4 Nov 2025 Nov 2025 Official leaderboard
51 Grok 4 Fast 141.5 138.9 to 143.2 Sep 2025 Sep 2025 Official leaderboard
52 Gemini 2.5 Pro Jun 2025 141.4 138.2 to 146.5 Jun 2025 Jun 2025 Official leaderboard
53 GPT-4.1 140.5 134.9 to 145 Apr 2025 Apr 2025 Official leaderboard
54 DeepSeek-V3.1 140.2 134 to 142.4 Aug 2025 Aug 2025 Official leaderboard
55 DeepSeek-R1 May 2025 140.0 135.1 to 142.2 May 2025 May 2025 Official leaderboard
56 Claude 3.7 Sonnet 139.8 136.6 to 142.5 Feb 2025 Feb 2025 Official leaderboard
57 o1 139.7 132.4 to 141.6 Dec 2024 Dec 2024 Official leaderboard
58 o3-mini 139.3 136 to 141.2 Jan 2025 Jan 2025 Official leaderboard
59 Gemini 3 Pro 138.0 131.7 to 145.5 Nov 2025 Nov 2025 Official leaderboard
60 GPT-4.5 137.5 133.4 to 139.6 Feb 2025 Feb 2025 Official leaderboard
61 Claude 3.5 Sonnet (snapshot unspecified) 137.3 133.9 to 139.8 Jun 2024 Official leaderboard
62 Muse Spark 137.3 130.6 to 144.5 Apr 2026 Apr 2026 Official leaderboard
63 DeepSeek-V3 (snapshot unspecified) 136.6 128.1 to 141.2 Dec 2024 Official leaderboard
64 GPT-4o Nov 2024 135.2 130.4 to 137.7 Nov 2024 Nov 2024 Official leaderboard
65 DeepSeek-R1 (snapshot unspecified) 134.0 112.1 to 137.5 Jan 2025 Official leaderboard
66 o1-mini 133.8 128.3 to 136.6 Sep 2024 Sep 2024 Official leaderboard
67 o1-preview 133.8 128.3 to 136.6 Sep 2024 Sep 2024 Official leaderboard
68 Mixtral 8x22B 131.9 126.2 to 135.2 Apr 2024 Apr 2024 Official leaderboard
69 Gemini 1.5 Pro (snapshot unspecified) 131.9 126.2 to 135.2 May 2024 Official leaderboard
70 Llama 3.1-405B 131.9 126.2 to 135.2 Jul 2024 Jul 2024 Official leaderboard
71 Llama 3-70B 129.2 106.7 to 132.6 Apr 2024 Apr 2024 Official leaderboard
72 GPT-4o (snapshot unspecified) 126.7 120.8 to 133.6 May 2024 Official leaderboard
73 Claude 3 Opus 126.5 121.6 to 132.4 Feb 2024 Feb 2024 Official leaderboard
74 GPT-4 Turbo (snapshot unspecified) 124.5 119.1 to 129.2 Apr 2024 Official leaderboard
75 GPT-3.5 Turbo (snapshot unspecified) 112.2 89.6 to 119 Jun 2023 Official leaderboard
#ModelScore ±SettingReleasedRun onSource
1 Claude Opus 5.5 171.1 166.7 to 177.3 Sep 2026 Sep 2026 Official leaderboard
2 GPT-6 Astra 171.0 166.9 to 178 Sep 2026 Sep 2026 Official leaderboard
3 Claude Mythos 5.1 169.6 165.4 to 174.6 Sep 2026 Sep 2026 Official leaderboard
4 GPT-6.1 Sol 168.9 164.9 to 174.8 Sep 2026 Sep 2026 Official leaderboard
5 Claude Mythos 5 168.3 164.2 to 178.5 Jun 2026 Jun 2026 Official leaderboard
6 Claude Opus 5 167.5 163 to 171.6 Jul 2026 Jul 2026 Official leaderboard
7 Claude Sonnet 5.5 167.3 163.3 to 171.7 Sep 2026 Sep 2026 Official leaderboard
8 GPT-5.6 Sol 166.5 163 to 171.3 Jul 2026 Jul 2026 Official leaderboard
9 GPT-6 Sol 164.2 155 to 167.5 Sep 2026 Sep 2026 Official leaderboard
10 GPT-5.5-Cyber 163.8 160.5 to 169.9 Jun 2026 May 2026 Official leaderboard
11 GPT-5.5 162.2 159.1 to 166.8 Apr 2026 Apr 2026 Official leaderboard
12 GLM-5.3 160.1 155.8 to 167 Aug 2026 Aug 2026 Official leaderboard
13 Claude Haiku 5.5 158.9 144.6 to 161.8 Oct 2026 Oct 2026 Official leaderboard
14 Claude Opus 4.8 158.8 155.8 to 163.1 May 2026 May 2026 Official leaderboard
15 Claude Opus 4.6 158.0 155.8 to 160.4 Feb 2026 Feb 2026 Official leaderboard
16 Kimi K3 157.6 148 to 160.7 Jul 2026 Jul 2026 Official leaderboard
17 GPT-6 Luna 157.3 109.2 to 159.9 Sep 2026 Sep 2026 Official leaderboard
18 GPT-5.4 157.0 155 to 159.6 Mar 2026 Mar 2026 Official leaderboard
19 GPT-5.3 Codex 156.8 ±2.4 Feb 2026 Feb 2026 Official leaderboard
20 Claude Sonnet 5 155.5 135.3 to 158.4 Jun 2026 Jun 2026 Official leaderboard
21 Claude Opus 4.7 154.9 153 to 157 Apr 2026 Apr 2026 Official leaderboard
22 Claude Sonnet 4.6 154.2 149.9 to 157.5 Feb 2026 Feb 2026 Official leaderboard
23 GPT-5.2 Codex 154.2 151.7 to 155.9 Dec 2025 Dec 2025 Official leaderboard
24 GLM-5.2 154.1 ±2.4 Jun 2026 Jun 2026 Official leaderboard
25 GPT-5.1-Codex-Max 153.7 150 to 155.2 Nov 2025 Nov 2025 Official leaderboard
26 Gemini 3.1 Pro 153.6 150.2 to 156 Feb 2026 Feb 2026 Official leaderboard
27 GLM-5.1 152.9 144.5 to 155.3 Apr 2026 Apr 2026 Official leaderboard
28 Claude Opus 4.5 152.9 150.8 to 155.1 Nov 2025 Nov 2025 Official leaderboard
29 GPT-5.2 152.6 150.2 to 154.8 Dec 2025 Dec 2025 Official leaderboard
30 Kimi K2.6 152.6 148.3 to 155.6 Apr 2026 Apr 2026 Official leaderboard
31 Claude Haiku 4.5 150.8 125.7 to 153.9 Oct 2025 Oct 2025 Official leaderboard
32 DeepSeek-V4-Pro 150.3 148 to 152.4 Apr 2026 Apr 2026 Official leaderboard
33 GPT-5.4 Mini 150.2 145.9 to 153 Mar 2026 Mar 2026 Official leaderboard
34 GPT-5 149.4 147.3 to 150.3 Aug 2025 Aug 2025 Official leaderboard
35 Claude Sonnet 4.5 149.2 147.3 to 150.9 Sep 2025 Sep 2025 Official leaderboard
36 GPT-5.1 148.0 144.9 to 150.3 Nov 2025 Nov 2025 Official leaderboard
37 GLM-5 147.9 143.2 to 151.7 Feb 2026 Feb 2026 Official leaderboard
38 Claude Opus 4.1 145.8 143.2 to 148.6 Aug 2025 Aug 2025 Official leaderboard
39 o3 145.8 142.2 to 147.9 Apr 2025 Apr 2025 Official leaderboard
40 GPT-5 mini 145.2 142.7 to 146.9 Aug 2025 Aug 2025 Official leaderboard
41 Grok 4 144.8 142.6 to 146.2 Jul 2025 Jul 2025 Official leaderboard
42 Claude Opus 4 144.7 142.3 to 146.7 May 2025 May 2025 Official leaderboard
43 o4-mini 144.2 131.7 to 147.1 Apr 2025 Apr 2025 Official leaderboard
44 Grok 4.1 143.8 141.7 to 145.2 Nov 2025 Nov 2025 Official leaderboard
45 Claude Sonnet 4 143.8 141.3 to 147 May 2025 May 2025 Official leaderboard
46 gpt-oss-120b 143.1 139.7 to 145.3 Aug 2025 Aug 2025 Official leaderboard
47 Kimi K2 Thinking 142.7 139 to 145.4 Nov 2025 Nov 2025 Official leaderboard
48 Grok 4 Fast 141.5 138.9 to 143.2 Sep 2025 Sep 2025 Official leaderboard
49 Gemini 2.5 Pro Jun 2025 141.4 138.2 to 146.5 Jun 2025 Jun 2025 Official leaderboard
50 GPT-4.1 140.5 134.9 to 145 Apr 2025 Apr 2025 Official leaderboard
51 DeepSeek-V3.1 140.2 134 to 142.4 Aug 2025 Aug 2025 Official leaderboard
52 DeepSeek-R1 May 2025 140.0 135.1 to 142.2 May 2025 May 2025 Official leaderboard
53 Claude 3.7 Sonnet 139.8 136.6 to 142.5 Feb 2025 Feb 2025 Official leaderboard
54 o1 139.7 132.4 to 141.6 Dec 2024 Dec 2024 Official leaderboard
55 o3-mini 139.3 136 to 141.2 Jan 2025 Jan 2025 Official leaderboard
56 Gemini 3 Pro 138.0 131.7 to 145.5 Nov 2025 Nov 2025 Official leaderboard
57 GPT-4.5 137.5 133.4 to 139.6 Feb 2025 Feb 2025 Official leaderboard
58 Muse Spark 137.3 130.6 to 144.5 Apr 2026 Apr 2026 Official leaderboard
59 GPT-4o Nov 2024 135.2 130.4 to 137.7 Nov 2024 Nov 2024 Official leaderboard
60 o1-mini 133.8 128.3 to 136.6 Sep 2024 Sep 2024 Official leaderboard
61 o1-preview 133.8 128.3 to 136.6 Sep 2024 Sep 2024 Official leaderboard
62 Mixtral 8x22B 131.9 126.2 to 135.2 Apr 2024 Apr 2024 Official leaderboard
63 Llama 3.1-405B 131.9 126.2 to 135.2 Jul 2024 Jul 2024 Official leaderboard
64 Llama 3-70B 129.2 106.7 to 132.6 Apr 2024 Apr 2024 Official leaderboard
65 Claude 3 Opus 126.5 121.6 to 132.4 Feb 2024 Feb 2024 Official leaderboard

19 models have held number one on the official leaderboard since 2023

Lab claims count from the day the card came out, independent runs from the model’s release.

Anthropic Mistral AI OpenAI
Own leaderboard 19 leaders

Source: Official Cyber ECI leaderboard

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 the benchmark’s own leaderboard
120 140 160 180 2024 2025 2026 120 140 160 180 2024 2025 2026

Source: Official Cyber ECI leaderboard

No lab model card reports it yet

We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.

More in Other

All 69 in Other
  • Artificial Analysis Intelligence Index v5 In use 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. Claude Fable 5.2 64.6
  • 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%

Data audit

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (75 entries).
  • Pass Models scored 65 models we track, plus 10 other entries on its leaderboard.
  • Pass Newest score The newest is from October 2026, for Claude Haiku 5.5.
  • Pass Frontier models 4 of the top 10 have a score, led by GPT-6 Astra (#4).
  • Note Headroom Scored as an index, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on epoch.ai: 65 rows for models we track and 10 other entries at the last read.
  • Pass Last read November 21st, 2026.
  • Pass Read worked The last read worked.
Consistency 1 pass 1 note
  • Note Lab claims No model has both a lab number and an independent run.
  • Pass Known issues No published audit found problems.

How we check