Benchmark Almanac

LLM Agronomy Benchmark (Farmers Business Network edition)

Gemini 4 Argon 94.3% leads since September 2026
Not in model cards
Version 2024-08

Models answer agronomy questions from six groups, including questions from Farmers Business Network. Each question carries equal weight in the overall score.

The October 2026 leaderboard lists 110 model configurations. The overall score covers 425 text questions and excludes image questions.

Made by
gbstox
Released
August 2024
Kind
Other professions
Level
University
Scored
Checked automatically
Answers
Multiple choice
Tasks
Public, plus a private test set
Measures
Capability
Language
English
In and out
Text → Text
Interaction
Single turn
Items from
Written by experts
Run by
A host
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
October 2026
Scores refreshed
November 20th, 2026
107
Entries on its own leaderboard, 94 models we track

Gemini 4 Argon leads with 94.3% Gemini 4 Argon leads with 94.3%

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.

107 entries. Scroll sideways for the rest.

  1. 94.3 Gemini 4 Argon
  2. 93.9 Gemini 3.1 Pro
  3. 93.9 google/gemini-3.1-pro-preview-customtools
  4. 92.7 Gemini 3.8 Flash
  5. 92.2 Grok 4.5
  6. 92.2 GPT-6 Astra
  7. 92.2 GPT-6 Sol
  8. 92.2 Muse Wonder
  9. 92.0 Kimi K3
  10. 92.0 GPT-6.1 Sol
  11. 92.0 Muse Spark Open
  12. 91.8 GPT-5.5
  13. 91.5 Grok 4.7
  14. 91.5 Claude Opus 5.5
  15. 91.3 GPT-5.6 Sol
  16. 91.3 GPT-5.6 Sol Pro
  17. 91.3 Ember-1
  18. 91.1 DeepSeek V4 Flash 0731
  19. 90.8 GPT-5
  20. 90.8 GPT-5.3 Codex
  21. 90.8 Gemini 3.1 Flash Image
  22. 90.8 reflection/beam-1.0-501b-a23b
  23. 90.6 Qwen3.5-Plus
  24. 90.6 GPT-5.3 Instant
  25. 90.6 Muse Spark 1.2
  26. 90.6 Tencent Hy4 preview
  27. 90.3 GPT-5.6 Terra
  28. 90.3 nvidia/nemotron-3-ultra-550b-a55b
  29. 90.3 openai/gpt-5.6-terra-pro
  30. 90.3 GLM-5.3
  31. 90.3 Nemotron 3.5 Super VL
  32. 90.1 o3
  33. 90.1 GPT-5.2 Codex
  34. 90.1 GPT-5.4 Pro
  35. 90.1 GLM-5.3-Flash
  36. 90.1 GPT-6 Luna
  37. 90.1 Claude Haiku 5.5
  38. 89.7 GPT-5.2
  39. 89.7 GLM-5
  40. 89.7 Claude Opus 4.8
  41. 89.7 GLM-5.2
  42. 89.7 GPT-5.6 Luna
  43. 89.4 GLM-5-Turbo
  44. 89.2 Claude Opus 4.1
  45. 89.2 Kimi K2.5
  46. 89.2 Qwen3.8 Flash
  47. 89.2 DeepSeek V4.1 Flash
  48. 88.9 Qwen3.5-27B
  49. 88.9 MiMo-V2.6 Pro
  50. 88.7 Claude Opus 4
  51. 88.7 Gemini 2.5 Pro
  52. 88.7 MiniMax-M3
  53. 88.7 Qwen3.7-Plus
  54. 88.5 Inkling
  55. 88.2 mistralai/mistral-large-4-0
  56. 87.8 o4-mini
  57. 87.8 Claude Opus 4.5
  58. 87.8 Step 3.7 Flash
  59. 87.5 Claude Sonnet 5
  60. 87.5 sakana/fugu-max
  61. 87.1 Qwen3.5-Flash
  62. 87.1 Qwen3.5-122B-A10B
  63. 87.1 Mercury 2.5
  64. 87.1 MiMo-V2.6 Flash
  65. 87.1 nex-agi/nex-n2.5-pro
  66. 86.8 GPT-5 mini
  67. 86.8 MiniMax-M2.5
  68. 86.8 Gemini 3.1 Flash-Lite
  69. 86.3 Claude Sonnet 4
  70. 86.3 Qwen 3.8 27B
  71. 85.7 GPT-5.4
  72. 85.4 Tencent Hy3
  73. 84.9 Qwen3-Max
  74. 84.5 Claude Sonnet 5.5
  75. 84.0 Inkling-Small
  76. 83.8 Doubao-Seed-1.6
  77. 83.8 Seed2.0 Lite (0215)
  78. 83.8 Ling-3.0-flash
  79. 82.6 GPT-4o
  80. 82.1 Llama 4 Maverick
  81. 82.1 gpt-oss-120b
  82. 82.1 glm-4.7-flash
  83. 81.2 deepseek/deepseek-chat
  84. 80.9 Nemotron 3 Nano 30B-A3B
  85. 80.7 Mercury 2
  86. 80.5 Qwen3-235B-A22B
  87. 78.8 Seed2.0 Mini
  88. 78.6 Qwen3-Coder-Next
  89. 77.7 meta-llama/llama-3.1-70b-instruct
  90. 76.5 gpt-oss-20b
  91. 76.2 Muse Spark 1.1
  92. 75.3 Qwen3-8B
  93. 74.3 Llama 4 Scout
  94. 74.1 GPT-4o mini
  95. 73.9 bytedance-seed/seed-1.6-flash
  96. 73.7 Gemma 3 27B
  97. 69.7 Qwen3-32B
  98. 69.4 Solar Mini 4
  99. 66.1 Solar Pro 3
  100. 63.1 Claude Fable 5
  101. 62.6 GPT-3.5 Turbo
  102. 56.2 meta-llama/llama-3.1-8b-instruct
  103. 54.8 dhenu2-in-8b-preview
  104. 54.8 pratik/llama3-8b-dhenu-0.1
  105. 50.6 Gemma 3 4B
  106. 49.4 Gemini 3 Flash
  107. 20.2 MiniMax-M2-her
  1. 1 Gemini 4 Argon 94.3%
  2. 2 Gemini 3.1 Pro 93.9%
  3. 3 google/gemini-3.1-pro-preview-customtools 93.9%
  4. 4 Gemini 3.8 Flash 92.7%
  5. 5 Grok 4.5 92.2%
  6. 6 GPT-6 Astra 92.2%
  7. 7 GPT-6 Sol 92.2%
  8. 8 Muse Wonder 92.2%
  9. 9 Kimi K3 92%
  10. 10 GPT-6.1 Sol 92%
  11. 11 Muse Spark Open 92%
  12. 12 GPT-5.5 91.8%
  13. 13 Grok 4.7 91.5%
  14. 14 Claude Opus 5.5 91.5%
  15. 15 GPT-5.6 Sol 91.3%
  16. 16 GPT-5.6 Sol Pro 91.3%
  17. 17 Ember-1 91.3%
  18. 18 DeepSeek V4 Flash 0731 91.1%
  19. 19 GPT-5 90.8%
  20. 20 GPT-5.3 Codex 90.8%
  21. 21 Gemini 3.1 Flash Image 90.8%
  22. 22 reflection/beam-1.0-501b-a23b 90.8%
  23. 23 Qwen3.5-Plus 90.6%
  24. 24 GPT-5.3 Instant 90.6%
  25. 25 Muse Spark 1.2 90.6%
  26. 26 Tencent Hy4 preview 90.6%
  27. 27 GPT-5.6 Terra 90.3%
  28. 28 nvidia/nemotron-3-ultra-550b-a55b 90.3%
  29. 29 openai/gpt-5.6-terra-pro 90.3%
  30. 30 GLM-5.3 90.3%
  31. 31 Nemotron 3.5 Super VL 90.3%
  32. 32 o3 90.1%
  33. 33 GPT-5.2 Codex 90.1%
  34. 34 GPT-5.4 Pro 90.1%
  35. 35 GLM-5.3-Flash 90.1%
  36. 36 GPT-6 Luna 90.1%
  37. 37 Claude Haiku 5.5 90.1%
  38. 38 GPT-5.2 89.7%
  39. 39 GLM-5 89.7%
  40. 40 Claude Opus 4.8 89.7%
  41. 41 GLM-5.2 89.7%
  42. 42 GPT-5.6 Luna pro 89.7%
  43. 43 GLM-5-Turbo 89.4%
  44. 44 Claude Opus 4.1 89.2%
  45. 45 Kimi K2.5 89.2%
  46. 46 Qwen3.8 Flash 89.2%
  47. 47 DeepSeek V4.1 Flash 89.2%
  48. 48 Qwen3.5-27B 88.9%
  49. 49 MiMo-V2.6 Pro 88.9%
  50. 50 Claude Opus 4 88.7%
  51. 51 Gemini 2.5 Pro Jun 2025 88.7%
  52. 52 MiniMax-M3 88.7%
  53. 53 Qwen3.7-Plus 88.7%
  54. 54 Inkling 88.5%
  55. 55 mistralai/mistral-large-4-0 88.2%
  56. 56 o4-mini high 87.8%
  57. 57 Claude Opus 4.5 87.8%
  58. 58 Step 3.7 Flash 87.8%
  59. 59 Claude Sonnet 5 87.5%
  60. 60 sakana/fugu-max 87.5%
  61. 61 Qwen3.5-Flash 87.1%
  62. 62 Qwen3.5-122B-A10B 87.1%
  63. 63 Mercury 2.5 87.1%
  64. 64 MiMo-V2.6 Flash 87.1%
  65. 65 nex-agi/nex-n2.5-pro 87.1%
  66. 66 GPT-5 mini 86.8%
  67. 67 MiniMax-M2.5 86.8%
  68. 68 Gemini 3.1 Flash-Lite 86.8%
  69. 69 Claude Sonnet 4 86.3%
  70. 70 Qwen 3.8 27B 86.3%
  71. 71 GPT-5.4 85.7%
  72. 72 Tencent Hy3 85.4%
  73. 73 Qwen3-Max 84.9%
  74. 74 Claude Sonnet 5.5 84.5%
  75. 75 Inkling-Small 84%
  76. 76 Doubao-Seed-1.6 83.8%
  77. 77 Seed2.0 Lite (0215) 83.8%
  78. 78 Ling-3.0-flash 83.8%
  79. 79 GPT-4o Mar 2025 82.6%
  80. 80 Llama 4 Maverick 82.1%
  81. 81 gpt-oss-120b 82.1%
  82. 82 glm-4.7-flash 82.1%
  83. 83 deepseek/deepseek-chat 81.2%
  84. 84 Nemotron 3 Nano 30B-A3B 80.9%
  85. 85 Mercury 2 80.7%
  86. 86 Qwen3-235B-A22B 80.5%
  87. 87 Seed2.0 Mini 78.8%
  88. 88 Qwen3-Coder-Next 78.6%
  89. 89 meta-llama/llama-3.1-70b-instruct 77.7%
  90. 90 gpt-oss-20b 76.5%
  91. 91 Muse Spark 1.1 76.2%
  92. 92 Qwen3-8B 75.3%
  93. 93 Llama 4 Scout 74.3%
  94. 94 GPT-4o mini 74.1%
  95. 95 bytedance-seed/seed-1.6-flash 73.9%
  96. 96 Gemma 3 27B 73.7%
  97. 97 Qwen3-32B 69.7%
  98. 98 Solar Mini 4 69.4%
  99. 99 Solar Pro 3 66.1%
  100. 100 Claude Fable 5 63.1%
  101. 101 GPT-3.5 Turbo Jan 2024 62.6%
  102. 102 meta-llama/llama-3.1-8b-instruct 56.2%
  103. 103 dhenu2-in-8b-preview 54.8%
  104. 104 pratik/llama3-8b-dhenu-0.1 54.8%
  105. 105 Gemma 3 4B 50.6%
  106. 106 Gemini 3 Flash 49.4%
  107. 107 MiniMax-M2-her 20.2%

Source: Official LLM Agronomy Benchmark (Farmers Business Network edition) leaderboard

94 models. Scroll sideways for the rest.

  1. 94.3 Gemini 4 Argon
  2. 93.9 Gemini 3.1 Pro
  3. 92.7 Gemini 3.8 Flash
  4. 92.2 Grok 4.5
  5. 92.2 GPT-6 Astra
  6. 92.2 GPT-6 Sol
  7. 92.2 Muse Wonder
  8. 92.0 Kimi K3
  9. 92.0 GPT-6.1 Sol
  10. 92.0 Muse Spark Open
  11. 91.8 GPT-5.5
  12. 91.5 Grok 4.7
  13. 91.5 Claude Opus 5.5
  14. 91.3 GPT-5.6 Sol
  15. 91.3 GPT-5.6 Sol Pro
  16. 91.3 Ember-1
  17. 91.1 DeepSeek V4 Flash 0731
  18. 90.8 GPT-5
  19. 90.8 GPT-5.3 Codex
  20. 90.8 Gemini 3.1 Flash Image
  21. 90.6 Qwen3.5-Plus
  22. 90.6 GPT-5.3 Instant
  23. 90.6 Muse Spark 1.2
  24. 90.6 Tencent Hy4 preview
  25. 90.3 GPT-5.6 Terra
  26. 90.3 GLM-5.3
  27. 90.3 Nemotron 3.5 Super VL
  28. 90.1 o3
  29. 90.1 GPT-5.2 Codex
  30. 90.1 GPT-5.4 Pro
  31. 90.1 GLM-5.3-Flash
  32. 90.1 GPT-6 Luna
  33. 90.1 Claude Haiku 5.5
  34. 89.7 GPT-5.2
  35. 89.7 GLM-5
  36. 89.7 Claude Opus 4.8
  37. 89.7 GLM-5.2
  38. 89.7 GPT-5.6 Luna
  39. 89.4 GLM-5-Turbo
  40. 89.2 Claude Opus 4.1
  41. 89.2 Kimi K2.5
  42. 89.2 Qwen3.8 Flash
  43. 89.2 DeepSeek V4.1 Flash
  44. 88.9 Qwen3.5-27B
  45. 88.9 MiMo-V2.6 Pro
  46. 88.7 Claude Opus 4
  47. 88.7 Gemini 2.5 Pro
  48. 88.7 MiniMax-M3
  49. 88.7 Qwen3.7-Plus
  50. 88.5 Inkling
  51. 87.8 o4-mini
  52. 87.8 Claude Opus 4.5
  53. 87.8 Step 3.7 Flash
  54. 87.5 Claude Sonnet 5
  55. 87.1 Qwen3.5-Flash
  56. 87.1 Qwen3.5-122B-A10B
  57. 87.1 Mercury 2.5
  58. 87.1 MiMo-V2.6 Flash
  59. 86.8 GPT-5 mini
  60. 86.8 MiniMax-M2.5
  61. 86.8 Gemini 3.1 Flash-Lite
  62. 86.3 Claude Sonnet 4
  63. 86.3 Qwen 3.8 27B
  64. 85.7 GPT-5.4
  65. 85.4 Tencent Hy3
  66. 84.9 Qwen3-Max
  67. 84.5 Claude Sonnet 5.5
  68. 84.0 Inkling-Small
  69. 83.8 Doubao-Seed-1.6
  70. 83.8 Seed2.0 Lite (0215)
  71. 83.8 Ling-3.0-flash
  72. 82.6 GPT-4o
  73. 82.1 Llama 4 Maverick
  74. 82.1 gpt-oss-120b
  75. 82.1 glm-4.7-flash
  76. 80.9 Nemotron 3 Nano 30B-A3B
  77. 80.7 Mercury 2
  78. 80.5 Qwen3-235B-A22B
  79. 78.8 Seed2.0 Mini
  80. 78.6 Qwen3-Coder-Next
  81. 76.5 gpt-oss-20b
  82. 76.2 Muse Spark 1.1
  83. 75.3 Qwen3-8B
  84. 74.3 Llama 4 Scout
  85. 74.1 GPT-4o mini
  86. 73.7 Gemma 3 27B
  87. 69.7 Qwen3-32B
  88. 69.4 Solar Mini 4
  89. 66.1 Solar Pro 3
  90. 63.1 Claude Fable 5
  91. 62.6 GPT-3.5 Turbo
  92. 50.6 Gemma 3 4B
  93. 49.4 Gemini 3 Flash
  94. 20.2 MiniMax-M2-her
  1. 1 Gemini 4 Argon 94.3%
  2. 2 Gemini 3.1 Pro 93.9%
  3. 3 Gemini 3.8 Flash 92.7%
  4. 4 Grok 4.5 92.2%
  5. 5 GPT-6 Astra 92.2%
  6. 6 GPT-6 Sol 92.2%
  7. 7 Muse Wonder 92.2%
  8. 8 Kimi K3 92%
  9. 9 GPT-6.1 Sol 92%
  10. 10 Muse Spark Open 92%
  11. 11 GPT-5.5 91.8%
  12. 12 Grok 4.7 91.5%
  13. 13 Claude Opus 5.5 91.5%
  14. 14 GPT-5.6 Sol 91.3%
  15. 15 GPT-5.6 Sol Pro 91.3%
  16. 16 Ember-1 91.3%
  17. 17 DeepSeek V4 Flash 0731 91.1%
  18. 18 GPT-5 90.8%
  19. 19 GPT-5.3 Codex 90.8%
  20. 20 Gemini 3.1 Flash Image 90.8%
  21. 21 Qwen3.5-Plus 90.6%
  22. 22 GPT-5.3 Instant 90.6%
  23. 23 Muse Spark 1.2 90.6%
  24. 24 Tencent Hy4 preview 90.6%
  25. 25 GPT-5.6 Terra 90.3%
  26. 26 GLM-5.3 90.3%
  27. 27 Nemotron 3.5 Super VL 90.3%
  28. 28 o3 90.1%
  29. 29 GPT-5.2 Codex 90.1%
  30. 30 GPT-5.4 Pro 90.1%
  31. 31 GLM-5.3-Flash 90.1%
  32. 32 GPT-6 Luna 90.1%
  33. 33 Claude Haiku 5.5 90.1%
  34. 34 GPT-5.2 89.7%
  35. 35 GLM-5 89.7%
  36. 36 Claude Opus 4.8 89.7%
  37. 37 GLM-5.2 89.7%
  38. 38 GPT-5.6 Luna pro 89.7%
  39. 39 GLM-5-Turbo 89.4%
  40. 40 Claude Opus 4.1 89.2%
  41. 41 Kimi K2.5 89.2%
  42. 42 Qwen3.8 Flash 89.2%
  43. 43 DeepSeek V4.1 Flash 89.2%
  44. 44 Qwen3.5-27B 88.9%
  45. 45 MiMo-V2.6 Pro 88.9%
  46. 46 Claude Opus 4 88.7%
  47. 47 Gemini 2.5 Pro Jun 2025 88.7%
  48. 48 MiniMax-M3 88.7%
  49. 49 Qwen3.7-Plus 88.7%
  50. 50 Inkling 88.5%
  51. 51 o4-mini high 87.8%
  52. 52 Claude Opus 4.5 87.8%
  53. 53 Step 3.7 Flash 87.8%
  54. 54 Claude Sonnet 5 87.5%
  55. 55 Qwen3.5-Flash 87.1%
  56. 56 Qwen3.5-122B-A10B 87.1%
  57. 57 Mercury 2.5 87.1%
  58. 58 MiMo-V2.6 Flash 87.1%
  59. 59 GPT-5 mini 86.8%
  60. 60 MiniMax-M2.5 86.8%
  61. 61 Gemini 3.1 Flash-Lite 86.8%
  62. 62 Claude Sonnet 4 86.3%
  63. 63 Qwen 3.8 27B 86.3%
  64. 64 GPT-5.4 85.7%
  65. 65 Tencent Hy3 85.4%
  66. 66 Qwen3-Max 84.9%
  67. 67 Claude Sonnet 5.5 84.5%
  68. 68 Inkling-Small 84%
  69. 69 Doubao-Seed-1.6 83.8%
  70. 70 Seed2.0 Lite (0215) 83.8%
  71. 71 Ling-3.0-flash 83.8%
  72. 72 GPT-4o Mar 2025 82.6%
  73. 73 Llama 4 Maverick 82.1%
  74. 74 gpt-oss-120b 82.1%
  75. 75 glm-4.7-flash 82.1%
  76. 76 Nemotron 3 Nano 30B-A3B 80.9%
  77. 77 Mercury 2 80.7%
  78. 78 Qwen3-235B-A22B 80.5%
  79. 79 Seed2.0 Mini 78.8%
  80. 80 Qwen3-Coder-Next 78.6%
  81. 81 gpt-oss-20b 76.5%
  82. 82 Muse Spark 1.1 76.2%
  83. 83 Qwen3-8B 75.3%
  84. 84 Llama 4 Scout 74.3%
  85. 85 GPT-4o mini 74.1%
  86. 86 Gemma 3 27B 73.7%
  87. 87 Qwen3-32B 69.7%
  88. 88 Solar Mini 4 69.4%
  89. 89 Solar Pro 3 66.1%
  90. 90 Claude Fable 5 63.1%
  91. 91 GPT-3.5 Turbo Jan 2024 62.6%
  92. 92 Gemma 3 4B 50.6%
  93. 93 Gemini 3 Flash 49.4%
  94. 94 MiniMax-M2-her 20.2%

Source: Official LLM Agronomy Benchmark (Farmers Business Network edition) leaderboard

#EntryScoreSettingReleasedRun onSource
1 Gemini 4 Argon 94.3% Sep 2026 Oct 2026 Official leaderboard
2 Gemini 3.1 Pro 93.9% Feb 2026 Jul 2026 Official leaderboard
3 google/gemini-3.1-pro-preview-customtools closed weights 93.9% Jul 2026 Official leaderboard
4 Gemini 3.8 Flash 92.7% Sep 2026 Sep 2026 Official leaderboard
5 Grok 4.5 92.2% Jul 2026 Jul 2026 Official leaderboard
6 GPT-6 Astra 92.2% Sep 2026 Sep 2026 Official leaderboard
7 GPT-6 Sol 92.2% Sep 2026 Sep 2026 Official leaderboard
8 Muse Wonder 92.2% Oct 2026 Oct 2026 Official leaderboard
9 Kimi K3 92% Jul 2026 Jul 2026 Official leaderboard
10 GPT-6.1 Sol 92% Sep 2026 Oct 2026 Official leaderboard
11 Muse Spark Open 92% Oct 2026 Oct 2026 Official leaderboard
12 GPT-5.5 91.8% Apr 2026 Jul 2026 Official leaderboard
13 Grok 4.7 91.5% Sep 2026 Sep 2026 Official leaderboard
14 Claude Opus 5.5 91.5% Sep 2026 Sep 2026 Official leaderboard
15 GPT-5.6 Sol 91.3% Jul 2026 Jul 2026 Official leaderboard
16 GPT-5.6 Sol Pro 91.3% Jul 2026 Jul 2026 Official leaderboard
17 Ember-1 91.3% Sep 2026 Oct 2026 Official leaderboard
18 DeepSeek V4 Flash 0731 91.1% Jul 2026 Aug 2026 Official leaderboard
19 GPT-5 90.8% Aug 2025 Jul 2026 Official leaderboard
20 GPT-5.3 Codex 90.8% Feb 2026 Jul 2026 Official leaderboard
21 Gemini 3.1 Flash Image 90.8% Feb 2026 Jul 2026 Official leaderboard
22 reflection/beam-1.0-501b-a23b open weights 90.8% Oct 2026 Official leaderboard
23 Qwen3.5-Plus 90.6% Feb 2026 Jul 2026 Official leaderboard
24 GPT-5.3 Instant 90.6% Mar 2026 Jul 2026 Official leaderboard
25 Muse Spark 1.2 90.6% Aug 2026 Aug 2026 Official leaderboard
26 Tencent Hy4 preview 90.6% Aug 2026 Sep 2026 Official leaderboard
27 GPT-5.6 Terra 90.3% Jul 2026 Jul 2026 Official leaderboard
28 nvidia/nemotron-3-ultra-550b-a55b open weights 90.3% Jul 2026 Official leaderboard
29 openai/gpt-5.6-terra-pro closed weights 90.3% Jul 2026 Official leaderboard
30 GLM-5.3 90.3% Aug 2026 Sep 2026 Official leaderboard
31 Nemotron 3.5 Super VL 90.3% Oct 2026 Oct 2026 Official leaderboard
32 o3 90.1% Apr 2025 Jul 2026 Official leaderboard
33 GPT-5.2 Codex 90.1% Dec 2025 Jul 2026 Official leaderboard
34 GPT-5.4 Pro 90.1% Mar 2026 Jul 2026 Official leaderboard
35 GLM-5.3-Flash 90.1% Aug 2026 Sep 2026 Official leaderboard
36 GPT-6 Luna 90.1% Sep 2026 Sep 2026 Official leaderboard
37 Claude Haiku 5.5 90.1% Oct 2026 Oct 2026 Official leaderboard
38 GPT-5.2 89.7% Dec 2025 Jul 2026 Official leaderboard
39 GLM-5 89.7% Feb 2026 Jul 2026 Official leaderboard
40 Claude Opus 4.8 89.7% May 2026 Jul 2026 Official leaderboard
41 GLM-5.2 89.7% Jun 2026 Jul 2026 Official leaderboard
42 GPT-5.6 Luna 89.7% pro Jul 2026 Jul 2026 Official leaderboard
43 GLM-5-Turbo 89.4% Mar 2026 Jul 2026 Official leaderboard
44 Claude Opus 4.1 89.2% Aug 2025 Jul 2026 Official leaderboard
45 Kimi K2.5 89.2% Jan 2026 Jul 2026 Official leaderboard
46 Qwen3.8 Flash 89.2% Aug 2026 Sep 2026 Official leaderboard
47 DeepSeek V4.1 Flash 89.2% Sep 2026 Sep 2026 Official leaderboard
48 Qwen3.5-27B 88.9% Feb 2026 Jul 2026 Official leaderboard
49 MiMo-V2.6 Pro 88.9% Sep 2026 Sep 2026 Official leaderboard
50 Claude Opus 4 88.7% May 2025 Jul 2026 Official leaderboard
51 Gemini 2.5 Pro Jun 2025 88.7% Jun 2025 Jul 2026 Official leaderboard
52 MiniMax-M3 88.7% Jun 2026 Jul 2026 Official leaderboard
53 Qwen3.7-Plus 88.7% Jun 2026 Jul 2026 Official leaderboard
54 Inkling 88.5% Jul 2026 Jul 2026 Official leaderboard
55 mistralai/mistral-large-4-0 open weights 88.2% Oct 2026 Official leaderboard
56 o4-mini 87.8% high Apr 2025 Jul 2026 Official leaderboard
57 Claude Opus 4.5 87.8% Nov 2025 Jul 2026 Official leaderboard
58 Step 3.7 Flash 87.8% May 2026 Aug 2026 Official leaderboard
59 Claude Sonnet 5 87.5% Jun 2026 Jul 2026 Official leaderboard
60 sakana/fugu-max closed weights 87.5% Sep 2026 Official leaderboard
61 Qwen3.5-Flash 87.1% Feb 2026 Jul 2026 Official leaderboard
62 Qwen3.5-122B-A10B 87.1% Feb 2026 Jul 2026 Official leaderboard
63 Mercury 2.5 87.1% Sep 2026 Sep 2026 Official leaderboard
64 MiMo-V2.6 Flash 87.1% Sep 2026 Sep 2026 Official leaderboard
65 nex-agi/nex-n2.5-pro open weights 87.1% Sep 2026 Official leaderboard
66 GPT-5 mini 86.8% Aug 2025 Jul 2026 Official leaderboard
67 MiniMax-M2.5 86.8% Feb 2026 Jul 2026 Official leaderboard
68 Gemini 3.1 Flash-Lite 86.8% Mar 2026 Jul 2026 Official leaderboard
69 Claude Sonnet 4 86.3% May 2025 Jul 2026 Official leaderboard
70 Qwen 3.8 27B 86.3% Aug 2026 Aug 2026 Official leaderboard
71 GPT-5.4 85.7% Mar 2026 Jul 2026 Official leaderboard
72 Tencent Hy3 85.4% Jul 2026 Jul 2026 Official leaderboard
73 Qwen3-Max 84.9% Sep 2025 Jul 2026 Official leaderboard
74 Claude Sonnet 5.5 84.5% Sep 2026 Oct 2026 Official leaderboard
75 Inkling-Small 84% Jul 2026 Jul 2026 Official leaderboard
76 Doubao-Seed-1.6 83.8% Jun 2025 Jul 2026 Official leaderboard
77 Seed2.0 Lite (0215) 83.8% Feb 2026 Jul 2026 Official leaderboard
78 Ling-3.0-flash 83.8% Aug 2026 Aug 2026 Official leaderboard
79 GPT-4o Mar 2025 82.6% Mar 2025 Jul 2026 Official leaderboard
80 Llama 4 Maverick 82.1% Apr 2025 Jul 2026 Official leaderboard
81 gpt-oss-120b 82.1% Aug 2025 Jul 2026 Official leaderboard
82 glm-4.7-flash 82.1% Jan 2026 Jul 2026 Official leaderboard
83 deepseek/deepseek-chat open weights 81.2% Jul 2026 Official leaderboard
84 Nemotron 3 Nano 30B-A3B 80.9% Dec 2025 Jul 2026 Official leaderboard
85 Mercury 2 80.7% Feb 2026 Jul 2026 Official leaderboard
86 Qwen3-235B-A22B 80.5% Apr 2025 Jul 2026 Official leaderboard
87 Seed2.0 Mini 78.8% Feb 2026 Jul 2026 Official leaderboard
88 Qwen3-Coder-Next 78.6% Feb 2026 Jul 2026 Official leaderboard
89 meta-llama/llama-3.1-70b-instruct open weights 77.7% Jul 2026 Official leaderboard
90 gpt-oss-20b 76.5% Aug 2025 Jul 2026 Official leaderboard
91 Muse Spark 1.1 76.2% Jul 2026 Jul 2026 Official leaderboard
92 Qwen3-8B 75.3% Apr 2025 Jul 2026 Official leaderboard
93 Llama 4 Scout 74.3% Apr 2025 Jul 2026 Official leaderboard
94 GPT-4o mini 74.1% Jul 2024 Jul 2026 Official leaderboard
95 bytedance-seed/seed-1.6-flash closed weights 73.9% Jul 2026 Official leaderboard
96 Gemma 3 27B 73.7% Mar 2025 Jul 2026 Official leaderboard
97 Qwen3-32B 69.7% Apr 2025 Jul 2026 Official leaderboard
98 Solar Mini 4 69.4% Oct 2026 Sep 2026 Official leaderboard
99 Solar Pro 3 66.1% Jan 2026 Jul 2026 Official leaderboard
100 Claude Fable 5 63.1% Jun 2026 Jul 2026 Official leaderboard
101 GPT-3.5 Turbo Jan 2024 62.6% Jan 2024 Jul 2026 Official leaderboard
102 meta-llama/llama-3.1-8b-instruct open weights 56.2% Jul 2026 Official leaderboard
103 dhenu2-in-8b-preview open weights 54.8% Jul 2026 Official leaderboard
104 pratik/llama3-8b-dhenu-0.1 open weights 54.8% Jul 2026 Official leaderboard
105 Gemma 3 4B 50.6% Mar 2025 Jul 2026 Official leaderboard
106 Gemini 3 Flash 49.4% Dec 2025 Jul 2026 Official leaderboard
107 MiniMax-M2-her 20.2% Jan 2026 Jul 2026 Official leaderboard
#ModelScoreSettingReleasedRun onSource
1 Gemini 4 Argon 94.3% Sep 2026 Oct 2026 Official leaderboard
2 Gemini 3.1 Pro 93.9% Feb 2026 Jul 2026 Official leaderboard
3 Gemini 3.8 Flash 92.7% Sep 2026 Sep 2026 Official leaderboard
4 Grok 4.5 92.2% Jul 2026 Jul 2026 Official leaderboard
5 GPT-6 Astra 92.2% Sep 2026 Sep 2026 Official leaderboard
6 GPT-6 Sol 92.2% Sep 2026 Sep 2026 Official leaderboard
7 Muse Wonder 92.2% Oct 2026 Oct 2026 Official leaderboard
8 Kimi K3 92% Jul 2026 Jul 2026 Official leaderboard
9 GPT-6.1 Sol 92% Sep 2026 Oct 2026 Official leaderboard
10 Muse Spark Open 92% Oct 2026 Oct 2026 Official leaderboard
11 GPT-5.5 91.8% Apr 2026 Jul 2026 Official leaderboard
12 Grok 4.7 91.5% Sep 2026 Sep 2026 Official leaderboard
13 Claude Opus 5.5 91.5% Sep 2026 Sep 2026 Official leaderboard
14 GPT-5.6 Sol 91.3% Jul 2026 Jul 2026 Official leaderboard
15 GPT-5.6 Sol Pro 91.3% Jul 2026 Jul 2026 Official leaderboard
16 Ember-1 91.3% Sep 2026 Oct 2026 Official leaderboard
17 DeepSeek V4 Flash 0731 91.1% Jul 2026 Aug 2026 Official leaderboard
18 GPT-5 90.8% Aug 2025 Jul 2026 Official leaderboard
19 GPT-5.3 Codex 90.8% Feb 2026 Jul 2026 Official leaderboard
20 Gemini 3.1 Flash Image 90.8% Feb 2026 Jul 2026 Official leaderboard
21 Qwen3.5-Plus 90.6% Feb 2026 Jul 2026 Official leaderboard
22 GPT-5.3 Instant 90.6% Mar 2026 Jul 2026 Official leaderboard
23 Muse Spark 1.2 90.6% Aug 2026 Aug 2026 Official leaderboard
24 Tencent Hy4 preview 90.6% Aug 2026 Sep 2026 Official leaderboard
25 GPT-5.6 Terra 90.3% Jul 2026 Jul 2026 Official leaderboard
26 GLM-5.3 90.3% Aug 2026 Sep 2026 Official leaderboard
27 Nemotron 3.5 Super VL 90.3% Oct 2026 Oct 2026 Official leaderboard
28 o3 90.1% Apr 2025 Jul 2026 Official leaderboard
29 GPT-5.2 Codex 90.1% Dec 2025 Jul 2026 Official leaderboard
30 GPT-5.4 Pro 90.1% Mar 2026 Jul 2026 Official leaderboard
31 GLM-5.3-Flash 90.1% Aug 2026 Sep 2026 Official leaderboard
32 GPT-6 Luna 90.1% Sep 2026 Sep 2026 Official leaderboard
33 Claude Haiku 5.5 90.1% Oct 2026 Oct 2026 Official leaderboard
34 GPT-5.2 89.7% Dec 2025 Jul 2026 Official leaderboard
35 GLM-5 89.7% Feb 2026 Jul 2026 Official leaderboard
36 Claude Opus 4.8 89.7% May 2026 Jul 2026 Official leaderboard
37 GLM-5.2 89.7% Jun 2026 Jul 2026 Official leaderboard
38 GPT-5.6 Luna 89.7% pro Jul 2026 Jul 2026 Official leaderboard
39 GLM-5-Turbo 89.4% Mar 2026 Jul 2026 Official leaderboard
40 Claude Opus 4.1 89.2% Aug 2025 Jul 2026 Official leaderboard
41 Kimi K2.5 89.2% Jan 2026 Jul 2026 Official leaderboard
42 Qwen3.8 Flash 89.2% Aug 2026 Sep 2026 Official leaderboard
43 DeepSeek V4.1 Flash 89.2% Sep 2026 Sep 2026 Official leaderboard
44 Qwen3.5-27B 88.9% Feb 2026 Jul 2026 Official leaderboard
45 MiMo-V2.6 Pro 88.9% Sep 2026 Sep 2026 Official leaderboard
46 Claude Opus 4 88.7% May 2025 Jul 2026 Official leaderboard
47 Gemini 2.5 Pro Jun 2025 88.7% Jun 2025 Jul 2026 Official leaderboard
48 MiniMax-M3 88.7% Jun 2026 Jul 2026 Official leaderboard
49 Qwen3.7-Plus 88.7% Jun 2026 Jul 2026 Official leaderboard
50 Inkling 88.5% Jul 2026 Jul 2026 Official leaderboard
51 o4-mini 87.8% high Apr 2025 Jul 2026 Official leaderboard
52 Claude Opus 4.5 87.8% Nov 2025 Jul 2026 Official leaderboard
53 Step 3.7 Flash 87.8% May 2026 Aug 2026 Official leaderboard
54 Claude Sonnet 5 87.5% Jun 2026 Jul 2026 Official leaderboard
55 Qwen3.5-Flash 87.1% Feb 2026 Jul 2026 Official leaderboard
56 Qwen3.5-122B-A10B 87.1% Feb 2026 Jul 2026 Official leaderboard
57 Mercury 2.5 87.1% Sep 2026 Sep 2026 Official leaderboard
58 MiMo-V2.6 Flash 87.1% Sep 2026 Sep 2026 Official leaderboard
59 GPT-5 mini 86.8% Aug 2025 Jul 2026 Official leaderboard
60 MiniMax-M2.5 86.8% Feb 2026 Jul 2026 Official leaderboard
61 Gemini 3.1 Flash-Lite 86.8% Mar 2026 Jul 2026 Official leaderboard
62 Claude Sonnet 4 86.3% May 2025 Jul 2026 Official leaderboard
63 Qwen 3.8 27B 86.3% Aug 2026 Aug 2026 Official leaderboard
64 GPT-5.4 85.7% Mar 2026 Jul 2026 Official leaderboard
65 Tencent Hy3 85.4% Jul 2026 Jul 2026 Official leaderboard
66 Qwen3-Max 84.9% Sep 2025 Jul 2026 Official leaderboard
67 Claude Sonnet 5.5 84.5% Sep 2026 Oct 2026 Official leaderboard
68 Inkling-Small 84% Jul 2026 Jul 2026 Official leaderboard
69 Doubao-Seed-1.6 83.8% Jun 2025 Jul 2026 Official leaderboard
70 Seed2.0 Lite (0215) 83.8% Feb 2026 Jul 2026 Official leaderboard
71 Ling-3.0-flash 83.8% Aug 2026 Aug 2026 Official leaderboard
72 GPT-4o Mar 2025 82.6% Mar 2025 Jul 2026 Official leaderboard
73 Llama 4 Maverick 82.1% Apr 2025 Jul 2026 Official leaderboard
74 gpt-oss-120b 82.1% Aug 2025 Jul 2026 Official leaderboard
75 glm-4.7-flash 82.1% Jan 2026 Jul 2026 Official leaderboard
76 Nemotron 3 Nano 30B-A3B 80.9% Dec 2025 Jul 2026 Official leaderboard
77 Mercury 2 80.7% Feb 2026 Jul 2026 Official leaderboard
78 Qwen3-235B-A22B 80.5% Apr 2025 Jul 2026 Official leaderboard
79 Seed2.0 Mini 78.8% Feb 2026 Jul 2026 Official leaderboard
80 Qwen3-Coder-Next 78.6% Feb 2026 Jul 2026 Official leaderboard
81 gpt-oss-20b 76.5% Aug 2025 Jul 2026 Official leaderboard
82 Muse Spark 1.1 76.2% Jul 2026 Jul 2026 Official leaderboard
83 Qwen3-8B 75.3% Apr 2025 Jul 2026 Official leaderboard
84 Llama 4 Scout 74.3% Apr 2025 Jul 2026 Official leaderboard
85 GPT-4o mini 74.1% Jul 2024 Jul 2026 Official leaderboard
86 Gemma 3 27B 73.7% Mar 2025 Jul 2026 Official leaderboard
87 Qwen3-32B 69.7% Apr 2025 Jul 2026 Official leaderboard
88 Solar Mini 4 69.4% Oct 2026 Sep 2026 Official leaderboard
89 Solar Pro 3 66.1% Jan 2026 Jul 2026 Official leaderboard
90 Claude Fable 5 63.1% Jun 2026 Jul 2026 Official leaderboard
91 GPT-3.5 Turbo Jan 2024 62.6% Jan 2024 Jul 2026 Official leaderboard
92 Gemma 3 4B 50.6% Mar 2025 Jul 2026 Official leaderboard
93 Gemini 3 Flash 49.4% Dec 2025 Jul 2026 Official leaderboard
94 MiniMax-M2-her 20.2% Jan 2026 Jul 2026 Official leaderboard

Rankings by part

We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Community Questions Fbn, Crop Management and Nutrient Management. Gemini 4 Argon leads 3 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

Community Questions Fbn

94 models we track have a score here.

  1. 1 GLM-5 ↑34 96%
  2. 2 Gemini 3.1 Pro 96%
  3. 3 GPT-5.6 Sol Pro ↑12 96%
  4. 4 Kimi K3 ↑4 96%
  5. 5 Gemini 3.8 Flash ↓2 96%
  6. 6 GPT-6 Astra ↓1 96%
  7. 7 Claude Opus 5.5 ↑6 96%
  8. 8 GPT-6 Sol ↓2 96%
  9. 9 Ember-1 ↑7 96%
  10. 10 GPT-6.1 Sol ↓1 96%

Scores at each effort level

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
DeepSeek V4 Flash 0731
max
91.1%
GLM-5.3-Flash
FlashX
90.1%
GPT-5.6 Luna
pro
89.7%

Source: Official LLM Agronomy Benchmark (Farmers Business Network edition) leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Gemini 4 Argon 94.3%
Gemini 3.1 Pro 93.9%
Gemini 3.8 Flash 92.7%
Grok 4.5 92.2%
GPT-6 Astra 92.2%
GPT-6 Sol 92.2%
Muse Wonder 92.2%
Kimi K3 92%
GPT-6.1 Sol 92%
Muse Spark Open 92%
GPT-5.5 91.8%
Grok 4.7 91.5%
Claude Opus 5.5 91.5%
GPT-5.6 Sol 91.3%
GPT-5.6 Sol Pro 91.3%
Ember-1 91.3%
DeepSeek V4 Flash 0731 91.1%
GPT-5 90.8%
GPT-5.3 Codex 90.8%
Gemini 3.1 Flash Image 90.8%
Qwen3.5-Plus 90.6%
GPT-5.3 Instant 90.6%
Muse Spark 1.2 90.6%
Tencent Hy4 preview 90.6%
GPT-5.6 Terra 90.3%
GLM-5.3 90.3%
Nemotron 3.5 Super VL 90.3%
o3 90.1%
GPT-5.2 Codex 90.1%
GPT-5.4 Pro 90.1%
GLM-5.3-Flash 90.1%
GPT-6 Luna 90.1%
Claude Haiku 5.5 90.1%
GPT-5.2 89.7%
GLM-5 89.7%
Claude Opus 4.8 89.7%
GLM-5.2 89.7%
GPT-5.6 Luna pro 89.7%
DeepSeek V4 Flash 0731 max 89.7%
GLM-5-Turbo 89.4%
GLM-5.3-Flash FlashX 89.4%
Claude Opus 4.1 89.2%
Kimi K2.5 89.2%
Qwen3.8 Flash 89.2%
DeepSeek V4.1 Flash 89.2%
Qwen3.5-27B 88.9%
MiMo-V2.6 Pro 88.9%
Claude Opus 4 88.7%
Gemini 2.5 Pro Jun 2025 88.7%
MiniMax-M3 88.7%
Qwen3.7-Plus 88.7%
Inkling 88.5%
GPT-5.6 Luna 88.2%
o4-mini high 87.8%
Claude Opus 4.5 87.8%
Step 3.7 Flash 87.8%
Claude Sonnet 5 87.5%
Qwen3.5-Flash 87.1%
Qwen3.5-122B-A10B 87.1%
Mercury 2.5 87.1%
MiMo-V2.6 Flash 87.1%
GPT-5 mini 86.8%
MiniMax-M2.5 86.8%
Gemini 3.1 Flash-Lite 86.8%
Claude Sonnet 4 86.3%
Qwen 3.8 27B 86.3%
GPT-5.4 85.7%
Tencent Hy3 85.4%
Qwen3-Max 84.9%
Claude Sonnet 5.5 84.5%
Inkling-Small 84%
Doubao-Seed-1.6 83.8%
Seed2.0 Lite (0215) 83.8%
Ling-3.0-flash 83.8%
GPT-4o Mar 2025 82.6%
Llama 4 Maverick 82.1%
gpt-oss-120b 82.1%
glm-4.7-flash 82.1%
Nemotron 3 Nano 30B-A3B 80.9%
Mercury 2 80.7%
Qwen3-235B-A22B 80.5%
Seed2.0 Mini 78.8%
Qwen3-Coder-Next 78.6%
gpt-oss-20b 76.5%
Muse Spark 1.1 76.2%
Qwen3-8B 75.3%
Llama 4 Scout 74.3%
GPT-4o mini 74.1%
Gemma 3 27B 73.7%
Qwen3-32B 69.7%
Solar Mini 4 69.4%
Solar Pro 3 66.1%
Claude Fable 5 63.1%
GPT-3.5 Turbo Jan 2024 62.6%
Gemma 3 4B 50.6%
Gemini 3 Flash 49.4%
MiniMax-M2-her 20.2%

Scores at each effort level

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
DeepSeek V4 Flash 0731
max
91.1%
GLM-5.3-Flash
FlashX
90.1%
GPT-5.6 Luna
pro
89.7%

Source: Official LLM Agronomy Benchmark (Farmers Business Network edition) leaderboard

ModelEffortHarnessScoreCost per taskOutput tokens
Gemini 4 Argon 94.3%
Gemini 3.1 Pro 93.9%
Gemini 3.8 Flash 92.7%
Grok 4.5 92.2%
GPT-6 Astra 92.2%
GPT-6 Sol 92.2%
Muse Wonder 92.2%
Kimi K3 92%
GPT-6.1 Sol 92%
Muse Spark Open 92%
GPT-5.5 91.8%
Grok 4.7 91.5%
Claude Opus 5.5 91.5%
GPT-5.6 Sol 91.3%
GPT-5.6 Sol Pro 91.3%
Ember-1 91.3%
DeepSeek V4 Flash 0731 91.1%
GPT-5 90.8%
GPT-5.3 Codex 90.8%
Gemini 3.1 Flash Image 90.8%
Qwen3.5-Plus 90.6%
GPT-5.3 Instant 90.6%
Muse Spark 1.2 90.6%
Tencent Hy4 preview 90.6%
GPT-5.6 Terra 90.3%
GLM-5.3 90.3%
Nemotron 3.5 Super VL 90.3%
o3 90.1%
GPT-5.2 Codex 90.1%
GPT-5.4 Pro 90.1%
GLM-5.3-Flash 90.1%
GPT-6 Luna 90.1%
Claude Haiku 5.5 90.1%
GPT-5.2 89.7%
GLM-5 89.7%
Claude Opus 4.8 89.7%
GLM-5.2 89.7%
GPT-5.6 Luna pro 89.7%
DeepSeek V4 Flash 0731 max 89.7%
GLM-5-Turbo 89.4%
GLM-5.3-Flash FlashX 89.4%
Claude Opus 4.1 89.2%
Kimi K2.5 89.2%
Qwen3.8 Flash 89.2%
DeepSeek V4.1 Flash 89.2%
Qwen3.5-27B 88.9%
MiMo-V2.6 Pro 88.9%
Claude Opus 4 88.7%
Gemini 2.5 Pro Jun 2025 88.7%
MiniMax-M3 88.7%
Qwen3.7-Plus 88.7%
Inkling 88.5%
GPT-5.6 Luna 88.2%
o4-mini high 87.8%
Claude Opus 4.5 87.8%
Step 3.7 Flash 87.8%
Claude Sonnet 5 87.5%
Qwen3.5-Flash 87.1%
Qwen3.5-122B-A10B 87.1%
Mercury 2.5 87.1%
MiMo-V2.6 Flash 87.1%
GPT-5 mini 86.8%
MiniMax-M2.5 86.8%
Gemini 3.1 Flash-Lite 86.8%
Claude Sonnet 4 86.3%
Qwen 3.8 27B 86.3%
GPT-5.4 85.7%
Tencent Hy3 85.4%
Qwen3-Max 84.9%
Claude Sonnet 5.5 84.5%
Inkling-Small 84%
Doubao-Seed-1.6 83.8%
Seed2.0 Lite (0215) 83.8%
Ling-3.0-flash 83.8%
GPT-4o Mar 2025 82.6%
Llama 4 Maverick 82.1%
gpt-oss-120b 82.1%
glm-4.7-flash 82.1%
Nemotron 3 Nano 30B-A3B 80.9%
Mercury 2 80.7%
Qwen3-235B-A22B 80.5%
Seed2.0 Mini 78.8%
Qwen3-Coder-Next 78.6%
gpt-oss-20b 76.5%
Muse Spark 1.1 76.2%
Qwen3-8B 75.3%
Llama 4 Scout 74.3%
GPT-4o mini 74.1%
Gemma 3 27B 73.7%
Qwen3-32B 69.7%
Solar Mini 4 69.4%
Solar Pro 3 66.1%
Claude Fable 5 63.1%
GPT-3.5 Turbo Jan 2024 62.6%
Gemma 3 4B 50.6%
Gemini 3 Flash 49.4%
MiniMax-M2-her 20.2%

6 models have held number one on the official leaderboard since its release in August 2024

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

OpenAI Google DeepMind
Own leaderboard Passed 85% in Apr 2025, by o3

Source: Official LLM Agronomy Benchmark (Farmers Business Network edition) leaderboard

O3 on the official leaderboard passed 85% in April 2025

That was 8 months after its release. By the same age, 17% of the benchmarks from 2024 had passed 85%. 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 Near the ceiling (85%) Gemini 4 Argon Gemini 3.1 Pro GPT-5 GPT-4o GPT-4o mini GPT-3.5 Turbo 0% 20% 40% 60% 80% 100% 2024 2025 2026 Near the ceiling (85%) Gemini 4 Argon GPT-4o

Source: Official LLM Agronomy Benchmark (Farmers Business Network edition) 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.

Versions

We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.

OpenAI Google DeepMind
LLM Agronomy Benchmark (Farmers Business Network edition) Passed 85% in Apr 2025, by o3
LLM Agronomy Benchmark (June 2024) Passed 85% in Jun 2024, by GPT-4o
GPT-4o 89.6%

Source: Official leaderboards

More in Law, finance and medicine

All 637 in Law, finance and medicine
  • MedXpertQA-MM In use Models answer expert-level medical exam questions that combine images with clinical records and test results. Scoring is accuracy. Qwen 3.8 Max 80.4%
  • HealthBench Professional In use 525 physician-authored tasks from real clinician chats (care consult, documentation, research) with physician-adjudicated rubrics. Claude Sonnet 5.5 69.2%
  • HealthBench In use 5,000 realistic health conversations graded against physician-written rubrics. A length-adjusted variant penalizes verbosity. GPT-5 67.2%
  • FinSearchComp In use Agents search the web to fetch live financial data, look up history, and run multi-step investigations, across 635 expert-written questions. Scored by answer accuracy. Ling-3.1-flash 81.7%
  • KorMedMCQA In use Models answer 7,469 multiple-choice questions from Korean licensing exams for doctors, nurses, pharmacists and dentists, held from 2012 to 2024. o1-preview 92.7%
  • Harvey LAB-AA v1.1 In use Artificial Analysis's run of Harvey's Legal Agent Benchmark on Harvey's newer private set of 120 tasks across 24 practice areas. A deliverable with a material hallucination scores zero, and the headline is the share of judges who find every rubric criterion passed. GPT-6.2 Astra 15%

Data audit

All 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access Public tasks, plus a private test set.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (107 entries).
  • Pass Models scored 94 models we track, plus 13 other entries on its leaderboard.
  • Pass Newest score The newest is from October 2026, for Muse Spark Open.
  • Pass Frontier models 5 of the top 10 have a score, led by GPT-6 Astra (#4).
  • Note Headroom Saturated. The best is 94.3%, past 85% since April 2025.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 97 rows for models we track and 13 other entries 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