Benchmark Almanac

Miðeind Icelandic Popular Culture

Claude Fable 5.2 39.5% leads since November 2026
Not in model cards

Models continue Icelandic song lyrics and identify the sources of familiar quotations. The two tasks measure word overlap and exact agreement, respectively.

Passes test 2. Miðeind reports scores for 60 named models on its current leaderboard. Miðeind counts the two tasks once in its average.

Made by
Miðeind
Released
September 2026
Kind
Knowledge in many languages
Level
Everyday
Scored
Checked automatically
Answers
Short answer
Tasks
Kept private
Measures
Capability
Language
Icelandic
In and out
Text → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
A host
Strongest model tested
Claude Fable 5.2 #2 on the capability index
Last model added
November 2026
Scores refreshed
November 20th, 2026
62
Entries on its own leaderboard, 61 models we track

Claude Fable 5.2 leads with 39.5% Claude Fable 5.2 leads with 39.5%

Best score per entry from the benchmark's own leaderboard, read on November 20th, 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.

  1. 1 Claude Fable 5.2 high 39.5%
  2. 2 Gemini 4 Argon high 29%
  3. 3 Gemini 4 Neon low 26%
  4. 4 Gemini 3.8 Flash high 24.4%
  5. 5 Gemini 3.7 Flash high 24.3%
  6. 6 Gemini 3.1 Pro high 24.2%
  7. 7 Claude Opus 5.5 high 24.1%
  8. 8 GPT-6.2 Sol high 22.7%
  9. 9 Claude Fable 5.1 high 22%
  10. 10 GPT-6 Astra high 22%
  11. 11 Gemini 4 Carbon low 21.7%
  12. 12 GPT-6.1 Sol high 21.6%
  13. 13 Claude Opus 5 high 21.3%
  14. 14 Gemini 3.5 Flash high 21%
  15. 15 Gemini 4 Hydrogen low 20.5%
  16. 16 GPT-5.6 Sol high 20%
  17. 17 GPT-6 Sol high 19.3%
  18. 18 GPT-6.1 Terra high 18%
  19. 19 Claude Sonnet 5.5 high 17.1%
  20. 20 GPT-5.2 high 15.5%
  21. 21 Claude Opus 4.8 high 15.4%
  22. 22 GPT-6 Luna high 15.2%
  23. 23 GPT-5.6 Luna high 14.9%
  24. 24 Gemini 2.5 Pro Jun 2025, high 14.8%
  25. 25 GPT-5.6 Terra high 14.4%
  26. 26 DeepSeek-V4-Pro on 13.3%
  27. 27 DeepSeek V4.1 Flash on 13.2%
  28. 28 Kimi K3 on 13%
  29. 29 Gemini 3.5 Flash-Lite high 12.8%
  30. 30 Inkling 2 Small minimal 12.6%
  31. 31 Inkling high 12.5%
  32. 32 Claude Haiku 5.5 high 12.1%
  33. 33 Claude Sonnet 5 high 11.8%
  34. 34 Qwen3.8-2.4T-A95B on 11.6%
  35. 35 GPT-4o Mar 2025, none 11%
  36. 36 GLM-5.3-Flash on 10.8%
  37. 37 Muse Spark Open high 10.7%
  38. 38 Gemma 4 31B IT on 8.92%
  39. 39 Grok 4.7 low 8.88%
  40. 40 Mistral Large 4 low 8.70%
  41. 41 Inkling-Small high 8.69%
  42. 42 Llama 3.1-70B none 8.55%
  43. 43 MiniMax-M3 on 8.39%
  44. 44 Gemma 4 26B A4B on 8.14%
  45. 45 Gemini 2.5 Flash-Lite Sep 2025, high 8.10%
  46. 46 MiniMax-M3.1-Flash high 7.80%
  47. 47 Mistral Medium 3.5 high 7.77%
  48. 48 Bielik 11B v3.1 Instruct off 7.76%
  49. 49 Claude Haiku 4.5 off 7.21%
  50. 50 GPT-3.5 Turbo Jan 2024, none 7.13%
  51. 51 Mistral Small 4 high 7.10%
  52. 52 Mistral Large 3 none 6.97%
  53. 53 Gemma 4 12B on 6.79%
  54. 54 Qwen 3.8 27B xhigh 6.66%
  55. 55 Gemma 4 E4B on 6.62%
  56. 56 Ripple-65B Preview on 6.48%
  57. 57 GPT-4.1 nano none 6.43%
  58. 58 Apertus 1.5 70B on 6.04%
  59. 59 Llama 3.1-8B none 5.84%
  60. 60 Gemma 4 E2B on 5.70%
  61. 61 Apertus 1.5 8B on 5.08%
  62. 62 Apertus 8B Instruct none 5.04%

Source: Official Miðeind Icelandic Popular Culture leaderboard

  1. 1 Claude Fable 5.2 high 39.5%
  2. 2 Gemini 4 Argon high 29%
  3. 3 Gemini 4 Neon low 26%
  4. 4 Gemini 3.8 Flash high 24.4%
  5. 5 Gemini 3.7 Flash high 24.3%
  6. 6 Gemini 3.1 Pro high 24.2%
  7. 7 Claude Opus 5.5 high 24.1%
  8. 8 GPT-6.2 Sol high 22.7%
  9. 9 Claude Fable 5.1 high 22%
  10. 10 GPT-6 Astra high 22%
  11. 11 Gemini 4 Carbon low 21.7%
  12. 12 GPT-6.1 Sol high 21.6%
  13. 13 Claude Opus 5 high 21.3%
  14. 14 Gemini 3.5 Flash high 21%
  15. 15 Gemini 4 Hydrogen low 20.5%
  16. 16 GPT-5.6 Sol high 20%
  17. 17 GPT-6 Sol high 19.3%
  18. 18 GPT-6.1 Terra high 18%
  19. 19 Claude Sonnet 5.5 high 17.1%
  20. 20 GPT-5.2 high 15.5%
  21. 21 Claude Opus 4.8 high 15.4%
  22. 22 GPT-6 Luna high 15.2%
  23. 23 GPT-5.6 Luna high 14.9%
  24. 24 Gemini 2.5 Pro Jun 2025, high 14.8%
  25. 25 GPT-5.6 Terra high 14.4%
  26. 26 DeepSeek-V4-Pro on 13.3%
  27. 27 DeepSeek V4.1 Flash on 13.2%
  28. 28 Kimi K3 on 13%
  29. 29 Gemini 3.5 Flash-Lite high 12.8%
  30. 30 Inkling 2 Small minimal 12.6%
  31. 31 Inkling high 12.5%
  32. 32 Claude Haiku 5.5 high 12.1%
  33. 33 Claude Sonnet 5 high 11.8%
  34. 34 Qwen3.8-2.4T-A95B on 11.6%
  35. 35 GPT-4o Mar 2025, none 11%
  36. 36 GLM-5.3-Flash on 10.8%
  37. 37 Muse Spark Open high 10.7%
  38. 38 Gemma 4 31B IT on 8.92%
  39. 39 Grok 4.7 low 8.88%
  40. 40 Mistral Large 4 low 8.70%
  41. 41 Inkling-Small high 8.69%
  42. 42 Llama 3.1-70B none 8.55%
  43. 43 MiniMax-M3 on 8.39%
  44. 44 Gemma 4 26B A4B on 8.14%
  45. 45 Gemini 2.5 Flash-Lite Sep 2025, high 8.10%
  46. 46 MiniMax-M3.1-Flash high 7.80%
  47. 47 Mistral Medium 3.5 high 7.77%
  48. 48 Claude Haiku 4.5 off 7.21%
  49. 49 GPT-3.5 Turbo Jan 2024, none 7.13%
  50. 50 Mistral Small 4 high 7.10%
  51. 51 Mistral Large 3 none 6.97%
  52. 52 Gemma 4 12B on 6.79%
  53. 53 Qwen 3.8 27B xhigh 6.66%
  54. 54 Gemma 4 E4B on 6.62%
  55. 55 Ripple-65B Preview on 6.48%
  56. 56 GPT-4.1 nano none 6.43%
  57. 57 Apertus 1.5 70B on 6.04%
  58. 58 Llama 3.1-8B none 5.84%
  59. 59 Gemma 4 E2B on 5.70%
  60. 60 Apertus 1.5 8B on 5.08%
  61. 61 Apertus 8B Instruct none 5.04%

Source: Official Miðeind Icelandic Popular Culture leaderboard

#EntryScoreSettingReleasedRun onLinkSource
1 Claude Fable 5.2 39.5% high Nov 2026 Nov 2026 anthropic.com Official leaderboard
2 Gemini 4 Argon 29% high Sep 2026 Oct 2026 ai.google.dev Official leaderboard
3 Gemini 4 Neon 26% low Nov 2026 Nov 2026 ai.google.dev Official leaderboard
4 Gemini 3.8 Flash 24.4% high Sep 2026 Sep 2026 ai.google.dev Official leaderboard
5 Gemini 3.7 Flash 24.3% high Aug 2026 ai.google.dev Official leaderboard
6 Gemini 3.1 Pro 24.2% high Feb 2026 ai.google.dev Official leaderboard
7 Claude Opus 5.5 24.1% high Sep 2026 Sep 2026 anthropic.com Official leaderboard
8 GPT-6.2 Sol 22.7% high Nov 2026 Nov 2026 developers.openai.com Official leaderboard
9 Claude Fable 5.1 22% high Sep 2026 anthropic.com Official leaderboard
10 GPT-6 Astra 22% high Sep 2026 developers.openai.com Official leaderboard
11 Gemini 4 Carbon 21.7% low Oct 2026 Nov 2026 ai.google.dev Official leaderboard
12 GPT-6.1 Sol 21.6% high Sep 2026 Sep 2026 developers.openai.com Official leaderboard
13 Claude Opus 5 21.3% high Jul 2026 anthropic.com Official leaderboard
14 Gemini 3.5 Flash 21% high May 2026 ai.google.dev Official leaderboard
15 Gemini 4 Hydrogen 20.5% low Nov 2026 Nov 2026 ai.google.dev Official leaderboard
16 GPT-5.6 Sol 20% high Jul 2026 developers.openai.com Official leaderboard
17 GPT-6 Sol 19.3% high Sep 2026 Sep 2026 developers.openai.com Official leaderboard
18 GPT-6.1 Terra 18% high Oct 2026 Oct 2026 developers.openai.com Official leaderboard
19 Claude Sonnet 5.5 17.1% high Sep 2026 Sep 2026 anthropic.com Official leaderboard
20 GPT-5.2 15.5% high Dec 2025 platform.openai.com Official leaderboard
21 Claude Opus 4.8 15.4% high May 2026 anthropic.com Official leaderboard
22 GPT-6 Luna 15.2% high Sep 2026 Sep 2026 developers.openai.com Official leaderboard
23 GPT-5.6 Luna 14.9% high Jul 2026 developers.openai.com Official leaderboard
24 Gemini 2.5 Pro Jun 2025 14.8% high Jun 2025 Sep 2026 ai.google.dev Official leaderboard
25 GPT-5.6 Terra 14.4% high Jul 2026 developers.openai.com Official leaderboard
26 DeepSeek-V4-Pro 13.3% on Apr 2026 Sep 2026 huggingface.co Official leaderboard
27 DeepSeek V4.1 Flash 13.2% on Sep 2026 Oct 2026 huggingface.co Official leaderboard
28 Kimi K3 13% on Jul 2026 huggingface.co Official leaderboard
29 Gemini 3.5 Flash-Lite 12.8% high Jul 2026 ai.google.dev Official leaderboard
30 Inkling 2 Small 12.6% minimal Nov 2026 Nov 2026 huggingface.co Official leaderboard
31 Inkling 12.5% high Jul 2026 Sep 2026 huggingface.co Official leaderboard
32 Claude Haiku 5.5 12.1% high Oct 2026 Oct 2026 anthropic.com Official leaderboard
33 Claude Sonnet 5 11.8% high Jun 2026 anthropic.com Official leaderboard
34 Qwen3.8-2.4T-A95B 11.6% on Aug 2026 Sep 2026 huggingface.co Official leaderboard
35 GPT-4o Mar 2025 11% none Mar 2025 Sep 2026 platform.openai.com Official leaderboard
36 GLM-5.3-Flash 10.8% on Aug 2026 Sep 2026 huggingface.co Official leaderboard
37 Muse Spark Open 10.7% high Oct 2026 Oct 2026 huggingface.co Official leaderboard
38 Gemma 4 31B IT 8.92% on Apr 2026 Sep 2026 deepmind.google Official leaderboard
39 Grok 4.7 8.88% low Sep 2026 Nov 2026 docs.x.ai Official leaderboard
40 Mistral Large 4 8.70% low Oct 2026 Oct 2026 mistral.ai Official leaderboard
41 Inkling-Small 8.69% high Jul 2026 Sep 2026 huggingface.co Official leaderboard
42 Llama 3.1-70B 8.55% none Jul 2024 Oct 2026 huggingface.co Official leaderboard
43 MiniMax-M3 8.39% on Jun 2026 huggingface.co Official leaderboard
44 Gemma 4 26B A4B 8.14% on Apr 2026 huggingface.co Official leaderboard
45 Gemini 2.5 Flash-Lite Sep 2025 8.10% high Sep 2025 ai.google.dev Official leaderboard
46 MiniMax-M3.1-Flash 7.80% high Oct 2026 Oct 2026 huggingface.co Official leaderboard
47 Mistral Medium 3.5 7.77% high Apr 2026 Sep 2026 mistral.ai Official leaderboard
48 Bielik 11B v3.1 Instruct 11.2B 7.76% off Nov 2026 huggingface.co Official leaderboard
49 Claude Haiku 4.5 7.21% off Oct 2025 Sep 2026 anthropic.com Official leaderboard
50 GPT-3.5 Turbo Jan 2024 7.13% none Jan 2024 Sep 2026 platform.openai.com Official leaderboard
51 Mistral Small 4 7.10% high Mar 2026 Sep 2026 mistral.ai Official leaderboard
52 Mistral Large 3 6.97% none Dec 2025 Sep 2026 mistral.ai Official leaderboard
53 Gemma 4 12B 6.79% on Jun 2026 Nov 2026 huggingface.co Official leaderboard
54 Qwen 3.8 27B 6.66% xhigh Aug 2026 huggingface.co Official leaderboard
55 Gemma 4 E4B 6.62% on Mar 2026 Oct 2026 huggingface.co Official leaderboard
56 Ripple-65B Preview 6.48% on Nov 2026 Nov 2026 huggingface.co Official leaderboard
57 GPT-4.1 nano 6.43% none Apr 2025 platform.openai.com Official leaderboard
58 Apertus 1.5 70B 6.04% on Jul 2026 Sep 2026 huggingface.co Official leaderboard
59 Llama 3.1-8B 5.84% none Jul 2024 Oct 2026 huggingface.co Official leaderboard
60 Gemma 4 E2B 5.70% on Mar 2026 Oct 2026 huggingface.co Official leaderboard
61 Apertus 1.5 8B 5.08% on Jul 2026 Sep 2026 huggingface.co Official leaderboard
62 Apertus 8B Instruct 5.04% none Sep 2025 Sep 2026 huggingface.co Official leaderboard
#ModelScoreSettingReleasedRun onLinkSource
1 Claude Fable 5.2 39.5% high Nov 2026 Nov 2026 anthropic.com Official leaderboard
2 Gemini 4 Argon 29% high Sep 2026 Oct 2026 ai.google.dev Official leaderboard
3 Gemini 4 Neon 26% low Nov 2026 Nov 2026 ai.google.dev Official leaderboard
4 Gemini 3.8 Flash 24.4% high Sep 2026 Sep 2026 ai.google.dev Official leaderboard
5 Gemini 3.7 Flash 24.3% high Aug 2026 ai.google.dev Official leaderboard
6 Gemini 3.1 Pro 24.2% high Feb 2026 ai.google.dev Official leaderboard
7 Claude Opus 5.5 24.1% high Sep 2026 Sep 2026 anthropic.com Official leaderboard
8 GPT-6.2 Sol 22.7% high Nov 2026 Nov 2026 developers.openai.com Official leaderboard
9 Claude Fable 5.1 22% high Sep 2026 anthropic.com Official leaderboard
10 GPT-6 Astra 22% high Sep 2026 developers.openai.com Official leaderboard
11 Gemini 4 Carbon 21.7% low Oct 2026 Nov 2026 ai.google.dev Official leaderboard
12 GPT-6.1 Sol 21.6% high Sep 2026 Sep 2026 developers.openai.com Official leaderboard
13 Claude Opus 5 21.3% high Jul 2026 anthropic.com Official leaderboard
14 Gemini 3.5 Flash 21% high May 2026 ai.google.dev Official leaderboard
15 Gemini 4 Hydrogen 20.5% low Nov 2026 Nov 2026 ai.google.dev Official leaderboard
16 GPT-5.6 Sol 20% high Jul 2026 developers.openai.com Official leaderboard
17 GPT-6 Sol 19.3% high Sep 2026 Sep 2026 developers.openai.com Official leaderboard
18 GPT-6.1 Terra 18% high Oct 2026 Oct 2026 developers.openai.com Official leaderboard
19 Claude Sonnet 5.5 17.1% high Sep 2026 Sep 2026 anthropic.com Official leaderboard
20 GPT-5.2 15.5% high Dec 2025 platform.openai.com Official leaderboard
21 Claude Opus 4.8 15.4% high May 2026 anthropic.com Official leaderboard
22 GPT-6 Luna 15.2% high Sep 2026 Sep 2026 developers.openai.com Official leaderboard
23 GPT-5.6 Luna 14.9% high Jul 2026 developers.openai.com Official leaderboard
24 Gemini 2.5 Pro Jun 2025 14.8% high Jun 2025 Sep 2026 ai.google.dev Official leaderboard
25 GPT-5.6 Terra 14.4% high Jul 2026 developers.openai.com Official leaderboard
26 DeepSeek-V4-Pro 13.3% on Apr 2026 Sep 2026 huggingface.co Official leaderboard
27 DeepSeek V4.1 Flash 13.2% on Sep 2026 Oct 2026 huggingface.co Official leaderboard
28 Kimi K3 13% on Jul 2026 huggingface.co Official leaderboard
29 Gemini 3.5 Flash-Lite 12.8% high Jul 2026 ai.google.dev Official leaderboard
30 Inkling 2 Small 12.6% minimal Nov 2026 Nov 2026 huggingface.co Official leaderboard
31 Inkling 12.5% high Jul 2026 Sep 2026 huggingface.co Official leaderboard
32 Claude Haiku 5.5 12.1% high Oct 2026 Oct 2026 anthropic.com Official leaderboard
33 Claude Sonnet 5 11.8% high Jun 2026 anthropic.com Official leaderboard
34 Qwen3.8-2.4T-A95B 11.6% on Aug 2026 Sep 2026 huggingface.co Official leaderboard
35 GPT-4o Mar 2025 11% none Mar 2025 Sep 2026 platform.openai.com Official leaderboard
36 GLM-5.3-Flash 10.8% on Aug 2026 Sep 2026 huggingface.co Official leaderboard
37 Muse Spark Open 10.7% high Oct 2026 Oct 2026 huggingface.co Official leaderboard
38 Gemma 4 31B IT 8.92% on Apr 2026 Sep 2026 deepmind.google Official leaderboard
39 Grok 4.7 8.88% low Sep 2026 Nov 2026 docs.x.ai Official leaderboard
40 Mistral Large 4 8.70% low Oct 2026 Oct 2026 mistral.ai Official leaderboard
41 Inkling-Small 8.69% high Jul 2026 Sep 2026 huggingface.co Official leaderboard
42 Llama 3.1-70B 8.55% none Jul 2024 Oct 2026 huggingface.co Official leaderboard
43 MiniMax-M3 8.39% on Jun 2026 huggingface.co Official leaderboard
44 Gemma 4 26B A4B 8.14% on Apr 2026 huggingface.co Official leaderboard
45 Gemini 2.5 Flash-Lite Sep 2025 8.10% high Sep 2025 ai.google.dev Official leaderboard
46 MiniMax-M3.1-Flash 7.80% high Oct 2026 Oct 2026 huggingface.co Official leaderboard
47 Mistral Medium 3.5 7.77% high Apr 2026 Sep 2026 mistral.ai Official leaderboard
48 Claude Haiku 4.5 7.21% off Oct 2025 Sep 2026 anthropic.com Official leaderboard
49 GPT-3.5 Turbo Jan 2024 7.13% none Jan 2024 Sep 2026 platform.openai.com Official leaderboard
50 Mistral Small 4 7.10% high Mar 2026 Sep 2026 mistral.ai Official leaderboard
51 Mistral Large 3 6.97% none Dec 2025 Sep 2026 mistral.ai Official leaderboard
52 Gemma 4 12B 6.79% on Jun 2026 Nov 2026 huggingface.co Official leaderboard
53 Qwen 3.8 27B 6.66% xhigh Aug 2026 huggingface.co Official leaderboard
54 Gemma 4 E4B 6.62% on Mar 2026 Oct 2026 huggingface.co Official leaderboard
55 Ripple-65B Preview 6.48% on Nov 2026 Nov 2026 huggingface.co Official leaderboard
56 GPT-4.1 nano 6.43% none Apr 2025 platform.openai.com Official leaderboard
57 Apertus 1.5 70B 6.04% on Jul 2026 Sep 2026 huggingface.co Official leaderboard
58 Llama 3.1-8B 5.84% none Jul 2024 Oct 2026 huggingface.co Official leaderboard
59 Gemma 4 E2B 5.70% on Mar 2026 Oct 2026 huggingface.co Official leaderboard
60 Apertus 1.5 8B 5.08% on Jul 2026 Sep 2026 huggingface.co Official leaderboard
61 Apertus 8B Instruct 5.04% none Sep 2025 Sep 2026 huggingface.co Official leaderboard

More effort adds a median 1 points

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 high
39.5%
Gemini 4 Argon
off high
29%
Gemini 3.8 Flash
low high
24.4%
Gemini 3.7 Flash
low high
24.3%
Gemini 3.1 Pro
low
24.2%
GPT-6.2 Sol
low high
22.7%
GPT-6 Astra
low
22%
GPT-6.1 Sol
low
21.6%
Claude Opus 5
off high
21.3%
Gemini 3.5 Flash
off high
21%
Gemini 4 Hydrogen
low high
20.5%
GPT-5.6 Sol
off high
20%

Source: Official Miðeind Icelandic Popular Culture leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Fable 5.2 high 39.5%
Claude Fable 5.2 low 37.4%
Gemini 4 Argon high 29%
Gemini 4 Neon low 26%
Gemini 3.8 Flash high 24.4%
Gemini 3.7 Flash high 24.3%
Gemini 3.1 Pro high 24.2%
Claude Opus 5.5 high 24.1%
Gemini 3.1 Pro low 23.4%
GPT-6.2 Sol high 22.7%
Claude Fable 5.1 high 22%
GPT-6 Astra high 22%
Gemini 3.7 Flash low 22%
Gemini 3.8 Flash low 21.9%
Gemini 4 Carbon low 21.7%
GPT-6.1 Sol high 21.6%
Claude Opus 5 high 21.3%
Gemini 3.5 Flash high 21%
GPT-6 Astra low 20.6%
Gemini 4 Hydrogen low 20.5%
Gemini 4 Argon off 20.5%
GPT-5.6 Sol high 20%
GPT-6.1 Sol low 19.5%
GPT-6 Sol high 19.3%
GPT-6.1 Terra high 18%
Gemini 4 Hydrogen high 17.2%
Claude Sonnet 5.5 high 17.1%
Claude Opus 5 off 17%
GPT-6.2 Sol low 16.9%
GPT-5.2 high 15.5%
Claude Opus 4.8 high 15.4%
GPT-6 Luna high 15.2%
GPT-5.6 Luna high 14.9%
Gemini 2.5 Pro Jun 2025 high 14.8%
GPT-5.6 Terra high 14.4%
Gemini 3.5 Flash off 14%
Gemini 2.5 Pro Jun 2025 minimal 13.8%
DeepSeek-V4-Pro on 13.3%
DeepSeek V4.1 Flash on 13.2%
Kimi K3 on 13%
Gemini 3.5 Flash-Lite high 12.8%
Inkling 2 Small minimal 12.6%
Inkling high 12.5%
Inkling 2 Small high 12.3%
GPT-6.1 Terra off 12.3%
GPT-5.6 Sol off 12.2%
Claude Haiku 5.5 high 12.1%
GPT-6 Sol off 11.9%
Claude Sonnet 5 high 11.8%
Claude Opus 4.8 off 11.8%
Qwen3.8-2.4T-A95B on 11.6%
Claude Sonnet 5.5 off 11.4%
Claude Sonnet 5 off 11.1%
GPT-4o Mar 2025 none 11%
GLM-5.3-Flash on 10.8%
Muse Spark Open high 10.7%
GPT-5.2 off 10.4%
DeepSeek-V4-Pro off 10.4%
Inkling off 10.1%
GPT-5.6 Terra off 10.1%
GPT-6 Luna off 9.94%
Kimi K3 off 9.91%
Gemini 3.5 Flash-Lite off 9.89%
DeepSeek V4.1 Flash off 9.58%
Claude Haiku 5.5 off 9.44%
GPT-5.6 Luna off 9.24%
Gemma 4 31B IT on 8.92%
Grok 4.7 low 8.88%
Mistral Large 4 low 8.70%
Inkling-Small high 8.69%
Mistral Large 4 high 8.65%
Llama 3.1-70B none 8.55%
MiniMax-M3 on 8.39%
Grok 4.7 high 8.26%
Gemma 4 26B A4B on 8.14%
Gemini 2.5 Flash-Lite Sep 2025 high 8.10%
Muse Spark Open off 7.85%
Gemini 2.5 Flash-Lite Sep 2025 off 7.81%
MiniMax-M3.1-Flash high 7.80%
Mistral Medium 3.5 high 7.77%
MiniMax-M3.1-Flash off 7.50%
Gemma 4 26B A4B off 7.24%
Claude Haiku 4.5 off 7.21%
GPT-3.5 Turbo Jan 2024 none 7.13%
Mistral Small 4 high 7.10%
Mistral Large 3 none 6.97%
Gemma 4 12B on 6.79%
Qwen 3.8 27B xhigh 6.66%
Claude Haiku 4.5 high 6.62%
Gemma 4 E4B on 6.62%
Gemma 4 12B off 6.52%
Ripple-65B Preview on 6.48%
GPT-4.1 nano none 6.43%
Qwen 3.8 27B off 6.34%
Gemma 4 E4B off 6.33%
Mistral Small 4 off 6.11%
Apertus 1.5 70B on 6.04%
Llama 3.1-8B none 5.84%
Mistral Medium 3.5 off 5.72%
Gemma 4 E2B on 5.70%
Gemma 4 E2B off 5.10%
Apertus 1.5 8B on 5.08%
Apertus 8B Instruct none 5.04%

More effort adds a median 1 points

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 high
39.5%
Gemini 4 Argon
off high
29%
Gemini 3.8 Flash
low high
24.4%
Gemini 3.7 Flash
low high
24.3%
Gemini 3.1 Pro
low
24.2%
GPT-6.2 Sol
low high
22.7%
GPT-6 Astra
low
22%
GPT-6.1 Sol
low
21.6%
Claude Opus 5
off high
21.3%
Gemini 3.5 Flash
off high
21%
Gemini 4 Hydrogen
low high
20.5%
GPT-5.6 Sol
off high
20%

Source: Official Miðeind Icelandic Popular Culture leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Claude Fable 5.2 high 39.5%
Claude Fable 5.2 low 37.4%
Gemini 4 Argon high 29%
Gemini 4 Neon low 26%
Gemini 3.8 Flash high 24.4%
Gemini 3.7 Flash high 24.3%
Gemini 3.1 Pro high 24.2%
Claude Opus 5.5 high 24.1%
Gemini 3.1 Pro low 23.4%
GPT-6.2 Sol high 22.7%
Claude Fable 5.1 high 22%
GPT-6 Astra high 22%
Gemini 3.7 Flash low 22%
Gemini 3.8 Flash low 21.9%
Gemini 4 Carbon low 21.7%
GPT-6.1 Sol high 21.6%
Claude Opus 5 high 21.3%
Gemini 3.5 Flash high 21%
GPT-6 Astra low 20.6%
Gemini 4 Hydrogen low 20.5%
Gemini 4 Argon off 20.5%
GPT-5.6 Sol high 20%
GPT-6.1 Sol low 19.5%
GPT-6 Sol high 19.3%
GPT-6.1 Terra high 18%
Gemini 4 Hydrogen high 17.2%
Claude Sonnet 5.5 high 17.1%
Claude Opus 5 off 17%
GPT-6.2 Sol low 16.9%
GPT-5.2 high 15.5%
Claude Opus 4.8 high 15.4%
GPT-6 Luna high 15.2%
GPT-5.6 Luna high 14.9%
Gemini 2.5 Pro Jun 2025 high 14.8%
GPT-5.6 Terra high 14.4%
Gemini 3.5 Flash off 14%
Gemini 2.5 Pro Jun 2025 minimal 13.8%
DeepSeek-V4-Pro on 13.3%
DeepSeek V4.1 Flash on 13.2%
Kimi K3 on 13%
Gemini 3.5 Flash-Lite high 12.8%
Inkling 2 Small minimal 12.6%
Inkling high 12.5%
Inkling 2 Small high 12.3%
GPT-6.1 Terra off 12.3%
GPT-5.6 Sol off 12.2%
Claude Haiku 5.5 high 12.1%
GPT-6 Sol off 11.9%
Claude Sonnet 5 high 11.8%
Claude Opus 4.8 off 11.8%
Qwen3.8-2.4T-A95B on 11.6%
Claude Sonnet 5.5 off 11.4%
Claude Sonnet 5 off 11.1%
GPT-4o Mar 2025 none 11%
GLM-5.3-Flash on 10.8%
Muse Spark Open high 10.7%
GPT-5.2 off 10.4%
DeepSeek-V4-Pro off 10.4%
Inkling off 10.1%
GPT-5.6 Terra off 10.1%
GPT-6 Luna off 9.94%
Kimi K3 off 9.91%
Gemini 3.5 Flash-Lite off 9.89%
DeepSeek V4.1 Flash off 9.58%
Claude Haiku 5.5 off 9.44%
GPT-5.6 Luna off 9.24%
Gemma 4 31B IT on 8.92%
Grok 4.7 low 8.88%
Mistral Large 4 low 8.70%
Inkling-Small high 8.69%
Mistral Large 4 high 8.65%
Llama 3.1-70B none 8.55%
MiniMax-M3 on 8.39%
Grok 4.7 high 8.26%
Gemma 4 26B A4B on 8.14%
Gemini 2.5 Flash-Lite Sep 2025 high 8.10%
Muse Spark Open off 7.85%
Gemini 2.5 Flash-Lite Sep 2025 off 7.81%
MiniMax-M3.1-Flash high 7.80%
Mistral Medium 3.5 high 7.77%
MiniMax-M3.1-Flash off 7.50%
Gemma 4 26B A4B off 7.24%
Claude Haiku 4.5 off 7.21%
GPT-3.5 Turbo Jan 2024 none 7.13%
Mistral Small 4 high 7.10%
Mistral Large 3 none 6.97%
Gemma 4 12B on 6.79%
Qwen 3.8 27B xhigh 6.66%
Claude Haiku 4.5 high 6.62%
Gemma 4 E4B on 6.62%
Gemma 4 12B off 6.52%
Ripple-65B Preview on 6.48%
GPT-4.1 nano none 6.43%
Qwen 3.8 27B off 6.34%
Gemma 4 E4B off 6.33%
Mistral Small 4 off 6.11%
Apertus 1.5 70B on 6.04%
Llama 3.1-8B none 5.84%
Mistral Medium 3.5 off 5.72%
Gemma 4 E2B on 5.70%
Gemma 4 E2B off 5.10%
Apertus 1.5 8B on 5.08%
Apertus 8B Instruct none 5.04%

4 models have held number one on the official leaderboard since its release in September 2026

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

Google DeepMind Anthropic
Own leaderboard 4 leaders

Source: Official Miðeind Icelandic Popular Culture 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 Best so far
0% 20% 40% 60% 80% 100% 2024 2025 2026 2027 Near the ceiling (85%) Claude Fable 5.2 Gemini 4 Argon Gemini 3.8 Flash Gemini 3.1 Pro GPT-5.2 Gemini 2.5 Pro Llama 3.1-70B 0% 20% 40% 60% 80% 100% 2024 2025 2026 2027 Near the ceiling (85%) Claude Fable 5.2 Gemini 4 Argon

Source: Official Miðeind Icelandic Popular Culture 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 Multilingual

All 545 in Multilingual
  • MMMLU In use MMLU professionally translated into 14 languages. Multiple-choice accuracy. Claude Mythos Preview 92.7%
  • WMT24++ In use English-to-55-language translation of WMT24 test data. Scored with automatic metrics such as MetricX or xCOMET. K-EXAONE 90.5%
  • MMLU-ProX In use MMLU-Pro translated into up to 29 languages with expert review. Multiple-choice accuracy per language. Qwen3.7-Max 87%
  • CLIcK In use Models answer 1,995 multiple-choice questions on Korean language and culture. Questions come from official Korean exams and textbooks in eleven categories. Solar Open 2 90.7%
  • Global MMLU Lite 3.0 In use Models answer 400 human-translated MMLU questions per language in 23 languages. Version 3.0 adds Czech, Hungarian, Oriya, Slovak and Tajik and updates Italian. Accuracy. Claude Opus 5 95.7%
  • MMMB In use Models answer multiple-choice questions about images in six languages across 15 categories. The set has about 12,000 questions. LFM2.5-VL-3B 83%

Data audit

All 11 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access The tasks are private.
  • Pass Confidence Checked against a primary source.
Scores 5 pass
  • Pass Independent runs Scores from the official leaderboard (62 entries).
  • Pass Models scored 61 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from November 2026, for Gemini 4 Neon.
  • Pass Frontier models 9 of the top 10 have a score, led by Claude Fable 5.2 (#2).
  • Pass Headroom The best is 39.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on mideind-icelandic-llm-leaderboard.static.hf.space: 103 rows for models we track and 1 other entry at the last read.
  • Pass Last read November 20th, 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