Benchmark Almanac

SnakeBench

GPT-5.1-Codex-Max 38 leads since Nov 2025 Not in model cards

Two language models each steer a snake on a shared board in a turn-based game, and the one that survives longest wins. Ratings come from many head-to-head games.

ARC Prize launched it as an experimental side quest with 2.8K games on 50 models. The live site now uses TrueSkill ratings and listed GPT-5.1-Codex-Max first on 2026-11-01.

Made by ARC Prize Foundation (Greg Kamradt) Released Feb 2025 Multi-agent games All tasks public Last model added Sep 2026 Scores refreshed 2 Nov 2026
Level
Everyday Most adults, without training
Scored
Ranked by matches Models play against each other, and the results become ratings.
Answers
Agent task Many steps in a terminal, browser, desktop, app, game or robot.
309
Entries on its own leaderboard, 183 models we track

GPT-5.1-Codex-Max leads with 38 among models released by Sep 2026

Best score per entry from the benchmark's own leaderboard, read on 2 Nov 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.

The newest model with a score here came out in Sep 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

309 entries. Scroll sideways for the rest.

  1. 38 GPT-5.1-Codex-Max
  2. 38 Gemini 3 Pro
  3. 38 GPT-6 Astra
  4. 37 GPT-5.5
  5. 37 Gemini 3.1 Pro
  6. 37 GPT-5.1-Codex-mini
  7. 36 o3
  8. 35 Claude Opus 4.8
  9. 35 GPT-5
  10. 35 Gemini 1.0 Pro
  11. 35 o4-mini
  12. 34 OpenAI GPT Latest
  13. 34 Qwen3.7-Max
  14. 34 Gemini 3.5 Flash
  15. 33 GPT-5.3 Codex
  16. 33 GPT-5-Codex
  17. 33 Gemini 2.0 Flash
  18. 33 Grok 4 Fast
  19. 33 Claude Opus 4.6
  20. 33 Claude Sonnet 4.5
  21. 33 GPT-5.1
  22. 33 GPT-5.2
  23. 33 Qwen3.5-122B-A10B
  24. 33 Qwen3 VL 30B A3B Thinking
  25. 32 GPT-5.2 Codex
  26. 32 Qwen3.5-Flash
  27. 32 GPT-4.1
  28. 32 Claude Sonnet 4
  29. 32 GPT-5.1-Codex
  30. 32 Mercury
  31. 32 Gemini 3 Flash
  32. 31 Gemini 2.5 Flash
  33. 31 DeepSeek V3.1 Terminus (exacto)
  34. 31 gpt-oss-safeguard-20b
  35. 31 Gemini 3.1 Pro Preview Custom Tools
  36. 31 Qwen3.5-27B
  37. 31 Qwen 3.6 Max (Preview)
  38. 31 Qwen3-Max
  39. 31 Gemini 3.1 Flash-Lite
  40. 31 GPT-5 Image
  41. 31 GLM-5
  42. 30 R1 Distill Llama 70B
  43. 30 Kimi K2 0905
  44. 30 Nano Banana (Gemini 2.5 Flash Image)
  45. 30 Claude Opus 4.5
  46. 30 Claude Opus 4.7
  47. 30 GPT-5.6 Terra
  48. 30 OpenAI deep research
  49. 29 Qwen3 235B A22B Instruct 2507
  50. 29 Claude Sonnet Latest
  51. 29 Gemini 2.5 Pro
  52. 29 Kimi K2 Thinking
  53. 29 GPT-5.4
  54. 29 GPT-5 Image Mini
  55. 29 Claude Opus Latest
  56. 29 Mercury Coder
  57. 29 Qwen3-VL-235B-A22B-Instruct
  58. 28 GPT Chat Latest
  59. 28 Codex Mini
  60. 28 Mercury 2
  61. 28 GLM-5V-Turbo
  62. 28 DeepSeek V4 Flash 0423
  63. 28 o3-mini
  64. 28 grok-code-fast-1
  65. 28 Qwen 3.6 Plus
  66. 28 GPT-5 nano
  67. 28 R1 Distill Qwen 14B
  68. 28 Qwen-VL-Max
  69. 27 Mistral Small 3.2 24B
  70. 27 KAT-Coder-Pro V2
  71. 27 Qwen3-Next-80B-A3B-Instruct
  72. 27 MiMo-V2.5-Pro
  73. 27 Claude 3.7 Sonnet
  74. 27 MiniMax-M2.5
  75. 27 Qwen3.5 Plus 2026-02-15
  76. 26 Claude Sonnet 4.6
  77. 26 Gemini 2.5 Flash Image Preview (nano-banana)
  78. 26 Qwen-Turbo
  79. 26 Qwen3-Next-80B-A3B
  80. 26 GLM-5-Turbo
  81. 26 Qwen3 30B A3B Thinking 2507
  82. 26 GPT-5 Chat
  83. 26 R1 0528
  84. 26 GLM-4.6V
  85. 26 DeepSeek R1T Chimera
  86. 26 GLM-5.1
  87. 26 Qwen Plus
  88. 26 gpt-oss-120b
  89. 26 GLM-4.7
  90. 26 Llama-3.1-Nemotron-70B-Instruct
  91. 26 Magistral Medium 1.0
  92. 26 Sonar Pro Search
  93. 26 Tongyi DeepResearch 30B A3B
  94. 26 GPT-5 mini
  95. 26 Claude Haiku 4.5
  96. 26 GPT-5.2 Instant
  97. 25 Devstral Small 1.1
  98. 25 ERNIE 4.5 VL 424B A47B
  99. 25 GPT-4o-mini (2024-07-18)
  100. 25 Olmo 3 7B Think
  101. 25 GLM-4.6
  102. 25 R1T Chimera
  103. 25 Gemini 2.5 Flash-Lite
  104. 25 Qwen3.5-35B-A3B
  105. 25 Qwen3-8B
  106. 25 Llama 3.2 90B
  107. 25 Mistral Small 3
  108. 25 GPT-4.1 mini
  109. 25 Doubao-Seed-1.6
  110. 25 qwen3-coder-plus
  111. 25 Qwen3-VL-235B-A22B
  112. 25 Ling-2.6-1T
  113. 25 Claude Opus 4.8 (Fast)
  114. 25 MiniMax-M3
  115. 25 Devstral 2 2512
  116. 25 QwQ 32B RpR v1 (free)
  117. 25 Gemma 3 27B
  118. 25 Kimi K2.5
  119. 25 GPT-5.6 Sol
  120. 24 DeepSeek V3 0324 (free)
  121. 24 Nemotron 3 Ultra
  122. 24 MiniMax-M1-80k
  123. 24 MiniMax-M2
  124. 24 Qwen 3.6 35B-A3B
  125. 24 DeepSeek V3.1 Nex N1
  126. 24 Qwen3-32B
  127. 24 Tencent Hy3 preview
  128. 24 DeepSeek R1 0528 Qwen3 8B (free)
  129. 24 GLM 4 32B
  130. 24 GLM-4.5-Air
  131. 24 o4 Mini Deep Research
  132. 24 Gemma 4 26B A4B
  133. 24 gpt-oss-120b (exacto)
  134. 24 Seed2.0 Lite (0215)
  135. 23 Claude Haiku Latest
  136. 23 DeepSeek-V3.1
  137. 23 Step 3.5 Flash
  138. 23 Qwen3.7-Plus
  139. 23 GPT-5.1 Chat
  140. 23 Aion-1.0
  141. 23 GPT-4o Search Preview
  142. 23 ERNIE-4.5-300B-A47B
  143. 23 Cydonia 24B V4.1
  144. 23 Llama 4 Maverick
  145. 23 Aion-1.0-Mini
  146. 23 Seed2.0 Mini
  147. 23 Qwen3-Coder-30B-A3B-Instruct
  148. 23 Ring-2.6-1T
  149. 23 Qwen Plus 0728 (thinking)
  150. 23 MiMo-V2-Flash
  151. 23 Qwen3.5 397B-A17B
  152. 22 Aion-2.0
  153. 22 MiniMax-M2.7
  154. 22 Olmo 3.1 32B Think
  155. 21 Mistral Small 3.1 24B
  156. 21 Mistral-7B-Instruct-v0.1
  157. 21 Qwen3 30B A3B Instruct 2507
  158. 21 Llama 3.3 70B Instruct
  159. 21 Gemini 2.0 Flash-Lite
  160. 21 Qwen3-Coder-480B-A35B
  161. 21 MiMo-V2.5
  162. 21 GPT-5.3 Instant
  163. 21 Step 3.7 Flash
  164. 21 Llama 3.3 Euryale 70B
  165. 21 Pixtral Large 2411
  166. 21 Palmyra X5
  167. 21 GLM-4.5V
  168. 21 DeepSeek-V3.1-Terminus
  169. 21 Hermes 3 405B Instruct
  170. 21 Llama 4 Scout
  171. 21 qwen3-coder-next
  172. 21 MiMo-V2-Pro
  173. 21 GPT-5.6 Luna
  174. 21 DeepSeek-V3
  175. 21 Hermes 4 405B
  176. 21 Qwen3-VL-8B-Thinking
  177. 20 Qwen Plus 0728
  178. 20 Qwen3-235B-A22B
  179. 20 MiMo-V2-Omni
  180. 20 Hermes 4 70B
  181. 20 Llama 3 70B Instruct
  182. 20 Perceptron Mk1
  183. 20 Qwen-Max
  184. 20 Phi-4-multimodal
  185. 20 Gemini Flash Latest
  186. 20 LongCat-Flash-Chat
  187. 20 DeepSeek-V3.2
  188. 20 GPT Mini Latest
  189. 20 Qwen2.5-Coder-32B-Instruct
  190. 20 Qwen2.5-VL-72B
  191. 20 Qwen3-14B
  192. 20 MiniMax-M2.1
  193. 20 GPT-4o mini
  194. 20 Trinity Large Thinking
  195. 19 InternVL3-78B
  196. 19 Olmo 3 32B Think
  197. 19 Gemma 2 9B
  198. 19 Gemma 3 12B
  199. 19 Magistral Small 2506
  200. 19 Sonar Pro
  201. 19 Ministral 8B
  202. 19 LlamaGuard 2 8B
  203. 19 Gemma 3n 4B
  204. 18 Qwen3 Coder 480B A35B (exacto)
  205. 18 ERNIE-4.5-21B-A3B-Thinking
  206. 18 Qwen3.6 27B
  207. 18 INTELLECT-3
  208. 18 Llama 3.1 8B Instruct
  209. 18 Llama Guard 4 12B
  210. 18 Qwen3 Coder Flash
  211. 18 OLMo 2 32B Instruct
  212. 18 KAT-Coder-Pro V1
  213. 18 Nemotron Nano 9B V2
  214. 18 Llama 3.2 11B
  215. 18 Command A
  216. 18 Voxtral Small
  217. 18 glm-4.7-flash
  218. 18 Gemma 4 31B IT
  219. 18 Jamba Large 1.7
  220. 18 Qwen2.5-72B
  221. 18 Phi-4
  222. 18 Step 3
  223. 18 Llama 3 Euryale 70B v2.1
  224. 18 Qwen2.5 Coder 7B Instruct
  225. 17 Qwen3.5 Plus 2026-04-20
  226. 17 Pixtral 12B
  227. 17 Nova Premier 1.0
  228. 17 deepseek-r1-0528-qwen3-8b
  229. 17 Qwen2.5-VL-32B
  230. 17 GLM-4.5
  231. 17 DeepSeek R1T2 Chimera
  232. 17 GPT-5.4 Image 2
  233. 17 Qwen3-235B-A22B-Thinking
  234. 17 Qwen 3.6 Flash
  235. 17 Virtuoso Large
  236. 16 Qwen3-4B
  237. 16 Kimi Latest
  238. 16 MAI DS R1
  239. 16 Seed 1.6 Flash
  240. 16 Qwen2.5 7B Instruct
  241. 16 WizardLM-2 8x22B
  242. 16 GPT-4.1 nano
  243. 16 Qwen3-30B-A3B
  244. 16 Nano Banana 2 (Gemini 3.1 Flash Image Preview)
  245. 16 Relace Apply 3
  246. 16 Ring-1T
  247. 16 MiniMax M2-her
  248. 15 Maestro Reasoning
  249. 15 Hunyuan-A13B-Instruct
  250. 15 LFM2-8B-A1B
  251. 15 Ling-1T
  252. 15 Skyfall 36B V2
  253. 15 Kimi-Dev-72B
  254. 15 GPT-5.4 Mini
  255. 15 Mistral NeMo
  256. 15 CodeLLaMa 7B Instruct Solidity
  257. 15 Kimi Linear 48B A3B Instruct
  258. 15 Mistral 7B Instruct v0.3
  259. 15 QwQ 32B RpR v1
  260. 15 Mistral-7B-Instruct-v0.2
  261. 14 Aion-RP 1.0 (8B)
  262. 14 Llama 3.2 3B Instruct
  263. 14 ERNIE 4.5 VL 28B A3B
  264. 14 Jamba Mini 1.7
  265. 14 Llama 3 8B Lunaris
  266. 14 UI-TARS 7B
  267. 14 Llama Guard 3 8B
  268. 14 Command R7B
  269. 13 Anubis 70B V1.1
  270. 13 GPT Audio
  271. 13 LFM2-2.6B
  272. 13 Mixtral 8x7B Instruct
  273. 13 Nova Lite 1.0
  274. 13 DeepHermes 3 Mistral 24B Preview
  275. 13 AFM-4.5B
  276. 13 GPT-4o Audio
  277. 13 MiniMax-01
  278. 13 MythoMax 13B
  279. 13 Qwen3 VL 8B Instruct
  280. 13 GPT-5.4 Nano
  281. 13 Morph V3 Large
  282. 13 GLM-4.1V-9B-Thinking
  283. 13 R1 (free)
  284. 12 Phi-3-mini-128k-instruct
  285. 12 GPT-4o-mini Search Preview
  286. 12 Granite 4.1 8B
  287. 12 Qianfan-OCR-Fast
  288. 12 Hermes 2 Pro - Llama-3 8B
  289. 11 Reka Flash 3
  290. 11 Llama 3 8B Instruct
  291. 10 Olmo 3 7B Instruct
  292. 10 ERNIE 4.5 21B A3B
  293. 10 Phi-3.5 Mini 128K Instruct
  294. 10 Phi 4 Mini Instruct
  295. 9 Nova Micro 1.0
  296. 9 DeepSeek-V3.2-Speciale
  297. 9 Llama 3.2 1B Instruct
  298. 9 Ministral 3B
  299. 9 DeepSeek V4 Pro 0423
  300. 9 Morph V3 Fast
  301. 9 Relace Search
  302. 9 GPT-4o
  303. 8 Qwen2.5-VL-7B
  304. 8 GPT Audio Mini
  305. 8 Gemma 3 4B
  306. 8 Spotlight
  307. 4 Reka Edge
  308. 1 Ling-2.6-flash
  309. 1 granite-4.0-micro
  1. 1 GPT-5.1-Codex-Max 38
  2. 2 Gemini 3 Pro 38
  3. 3 GPT-6 Astra 38
  4. 4 GPT-5.5 37
  5. 5 Gemini 3.1 Pro 37
  6. 6 GPT-5.1-Codex-mini 37
  7. 7 o3 36
  8. 8 Claude Opus 4.8 35
  9. 9 GPT-5 35
  10. 10 Gemini 1.0 Pro 35
  11. 11 o4-mini 35
  12. 12 OpenAI GPT Latest 34
  13. 13 Qwen3.7-Max 34
  14. 14 Gemini 3.5 Flash 34
  15. 15 GPT-5.3 Codex 33
  16. 16 GPT-5-Codex 33
  17. 17 Gemini 2.0 Flash Feb 2025 33
  18. 18 Grok 4 Fast 33
  19. 19 Claude Opus 4.6 33
  20. 20 Claude Sonnet 4.5 33
  21. 21 GPT-5.1 33
  22. 22 GPT-5.2 33
  23. 23 Qwen3.5-122B-A10B 33
  24. 24 Qwen3 VL 30B A3B Thinking 33
  25. 25 GPT-5.2 Codex 32
  26. 26 Qwen3.5-Flash 32
  27. 27 GPT-4.1 32
  28. 28 Claude Sonnet 4 32
  29. 29 GPT-5.1-Codex 32
  30. 30 Mercury 32
  31. 31 Gemini 3 Flash 32
  32. 32 Gemini 2.5 Flash Sep 2025 31
  33. 33 DeepSeek V3.1 Terminus (exacto) 31
  34. 34 gpt-oss-safeguard-20b 31
  35. 35 Gemini 3.1 Pro Preview Custom Tools 31
  36. 36 Qwen3.5-27B 31
  37. 37 Qwen 3.6 Max (Preview) 31
  38. 38 Qwen3-Max 31
  39. 39 Gemini 3.1 Flash-Lite 31
  40. 40 GPT-5 Image 31
  41. 41 GLM-5 31
  42. 42 R1 Distill Llama 70B 30
  43. 43 Kimi K2 0905 30
  44. 44 Nano Banana (Gemini 2.5 Flash Image) 30
  45. 45 Claude Opus 4.5 30
  46. 46 Claude Opus 4.7 30
  47. 47 GPT-5.6 Terra 30
  48. 48 OpenAI deep research 30
  49. 49 Qwen3 235B A22B Instruct 2507 29
  50. 50 Claude Sonnet Latest 29
  51. 51 Gemini 2.5 Pro Jun 2025 29
  52. 52 Kimi K2 Thinking 29
  53. 53 GPT-5.4 29
  54. 54 GPT-5 Image Mini 29
  55. 55 Claude Opus Latest 29
  56. 56 Mercury Coder 29
  57. 57 Qwen3-VL-235B-A22B-Instruct 29
  58. 58 GPT Chat Latest 28
  59. 59 Codex Mini 28
  60. 60 Mercury 2 28
  61. 61 GLM-5V-Turbo 28
  62. 62 DeepSeek V4 Flash 0423 28
  63. 63 o3-mini 28
  64. 64 grok-code-fast-1 28
  65. 65 Qwen 3.6 Plus 28
  66. 66 GPT-5 nano 28
  67. 67 R1 Distill Qwen 14B 28
  68. 68 Qwen-VL-Max 28
  69. 69 Mistral Small 3.2 24B 27
  70. 70 KAT-Coder-Pro V2 27
  71. 71 Qwen3-Next-80B-A3B-Instruct 27
  72. 72 MiMo-V2.5-Pro 27
  73. 73 Claude 3.7 Sonnet thinking 27
  74. 74 MiniMax-M2.5 27
  75. 75 Qwen3.5 Plus 2026-02-15 27
  76. 76 Claude Sonnet 4.6 26
  77. 77 Gemini 2.5 Flash Image Preview (nano-banana) 26
  78. 78 Qwen-Turbo 26
  79. 79 Qwen3-Next-80B-A3B thinking 26
  80. 80 GLM-5-Turbo 26
  81. 81 Qwen3 30B A3B Thinking 2507 26
  82. 82 GPT-5 Chat 26
  83. 83 R1 0528 26
  84. 84 GLM-4.6V 26
  85. 85 DeepSeek R1T Chimera 26
  86. 86 GLM-5.1 26
  87. 87 Qwen Plus Apr 2025 26
  88. 88 gpt-oss-120b 26
  89. 89 GLM-4.7 26
  90. 90 Llama-3.1-Nemotron-70B-Instruct 26
  91. 91 Magistral Medium 1.0 26
  92. 92 Sonar Pro Search 26
  93. 93 Tongyi DeepResearch 30B A3B 26
  94. 94 GPT-5 mini 26
  95. 95 Claude Haiku 4.5 26
  96. 96 GPT-5.2 Instant 26
  97. 97 Devstral Small 1.1 25
  98. 98 ERNIE 4.5 VL 424B A47B 25
  99. 99 GPT-4o-mini (2024-07-18) 25
  100. 100 Olmo 3 7B Think 25
  101. 101 GLM-4.6 25
  102. 102 R1T Chimera 25
  103. 103 Gemini 2.5 Flash-Lite Sep 2025 25
  104. 104 Qwen3.5-35B-A3B 25
  105. 105 Qwen3-8B 25
  106. 106 Llama 3.2 90B 25
  107. 107 Mistral Small 3 25
  108. 108 GPT-4.1 mini 25
  109. 109 Doubao-Seed-1.6 25
  110. 110 qwen3-coder-plus 25
  111. 111 Qwen3-VL-235B-A22B thinking 25
  112. 112 Ling-2.6-1T 25
  113. 113 Claude Opus 4.8 (Fast) 25
  114. 114 MiniMax-M3 25
  115. 115 Devstral 2 2512 25
  116. 116 QwQ 32B RpR v1 (free) 25
  117. 117 Gemma 3 27B 25
  118. 118 Kimi K2.5 25
  119. 119 GPT-5.6 Sol 25
  120. 120 DeepSeek V3 0324 (free) 24
  121. 121 Nemotron 3 Ultra 24
  122. 122 MiniMax-M1-80k 24
  123. 123 MiniMax-M2 24
  124. 124 Qwen 3.6 35B-A3B 24
  125. 125 DeepSeek V3.1 Nex N1 24
  126. 126 Qwen3-32B 24
  127. 127 Tencent Hy3 preview 24
  128. 128 DeepSeek R1 0528 Qwen3 8B (free) 24
  129. 129 GLM 4 32B 24
  130. 130 GLM-4.5-Air 24
  131. 131 o4 Mini Deep Research 24
  132. 132 Gemma 4 26B A4B 24
  133. 133 gpt-oss-120b (exacto) 24
  134. 134 Seed2.0 Lite (0215) 24
  135. 135 Claude Haiku Latest 23
  136. 136 DeepSeek-V3.1 23
  137. 137 Step 3.5 Flash 23
  138. 138 Qwen3.7-Plus 23
  139. 139 GPT-5.1 Chat 23
  140. 140 Aion-1.0 23
  141. 141 GPT-4o Search Preview 23
  142. 142 ERNIE-4.5-300B-A47B 23
  143. 143 Cydonia 24B V4.1 23
  144. 144 Llama 4 Maverick 23
  145. 145 Aion-1.0-Mini 23
  146. 146 Seed2.0 Mini 23
  147. 147 Qwen3-Coder-30B-A3B-Instruct 23
  148. 148 Ring-2.6-1T 23
  149. 149 Qwen Plus 0728 (thinking) 23
  150. 150 MiMo-V2-Flash 23
  151. 151 Qwen3.5 397B-A17B 23
  152. 152 Aion-2.0 22
  153. 153 MiniMax-M2.7 22
  154. 154 Olmo 3.1 32B Think 22
  155. 155 Mistral Small 3.1 24B 21
  156. 156 Mistral-7B-Instruct-v0.1 21
  157. 157 Qwen3 30B A3B Instruct 2507 21
  158. 158 Llama 3.3 70B Instruct 21
  159. 159 Gemini 2.0 Flash-Lite 21
  160. 160 Qwen3-Coder-480B-A35B 21
  161. 161 MiMo-V2.5 21
  162. 162 GPT-5.3 Instant 21
  163. 163 Step 3.7 Flash 21
  164. 164 Llama 3.3 Euryale 70B 21
  165. 165 Pixtral Large 2411 21
  166. 166 Palmyra X5 21
  167. 167 GLM-4.5V 21
  168. 168 DeepSeek-V3.1-Terminus 21
  169. 169 Hermes 3 405B Instruct 21
  170. 170 Llama 4 Scout 21
  171. 171 qwen3-coder-next 21
  172. 172 MiMo-V2-Pro 21
  173. 173 GPT-5.6 Luna 21
  174. 174 DeepSeek-V3 Mar 2025 21
  175. 175 Hermes 4 405B 21
  176. 176 Qwen3-VL-8B-Thinking 21
  177. 177 Qwen Plus 0728 20
  178. 178 Qwen3-235B-A22B 20
  179. 179 MiMo-V2-Omni 20
  180. 180 Hermes 4 70B 20
  181. 181 Llama 3 70B Instruct 20
  182. 182 Perceptron Mk1 20
  183. 183 Qwen-Max 20
  184. 184 Phi-4-multimodal 20
  185. 185 Gemini Flash Latest 20
  186. 186 LongCat-Flash-Chat 20
  187. 187 DeepSeek-V3.2 20
  188. 188 GPT Mini Latest 20
  189. 189 Qwen2.5-Coder-32B-Instruct 20
  190. 190 Qwen2.5-VL-72B 20
  191. 191 Qwen3-14B 20
  192. 192 MiniMax-M2.1 20
  193. 193 GPT-4o mini 20
  194. 194 Trinity Large Thinking 20
  195. 195 InternVL3-78B 19
  196. 196 Olmo 3 32B Think 19
  197. 197 Gemma 2 9B 19
  198. 198 Gemma 3 12B 19
  199. 199 Magistral Small 2506 19
  200. 200 Sonar Pro 19
  201. 201 Ministral 8B 19
  202. 202 LlamaGuard 2 8B 19
  203. 203 Gemma 3n 4B 19
  204. 204 Qwen3 Coder 480B A35B (exacto) 18
  205. 205 ERNIE-4.5-21B-A3B-Thinking 18
  206. 206 Qwen3.6 27B 18
  207. 207 INTELLECT-3 18
  208. 208 Llama 3.1 8B Instruct 18
  209. 209 Llama Guard 4 12B 18
  210. 210 Qwen3 Coder Flash 18
  211. 211 OLMo 2 32B Instruct 18
  212. 212 KAT-Coder-Pro V1 18
  213. 213 Nemotron Nano 9B V2 18
  214. 214 Llama 3.2 11B 18
  215. 215 Command A 18
  216. 216 Voxtral Small 18
  217. 217 glm-4.7-flash 18
  218. 218 Gemma 4 31B IT 18
  219. 219 Jamba Large 1.7 18
  220. 220 Qwen2.5-72B 18
  221. 221 Phi-4 18
  222. 222 Step 3 18
  223. 223 Llama 3 Euryale 70B v2.1 18
  224. 224 Qwen2.5 Coder 7B Instruct 18
  225. 225 Qwen3.5 Plus 2026-04-20 17
  226. 226 Pixtral 12B 17
  227. 227 Nova Premier 1.0 17
  228. 228 deepseek-r1-0528-qwen3-8b 17
  229. 229 Qwen2.5-VL-32B 17
  230. 230 GLM-4.5 17
  231. 231 DeepSeek R1T2 Chimera 17
  232. 232 GPT-5.4 Image 2 17
  233. 233 Qwen3-235B-A22B-Thinking Jul 2025 17
  234. 234 Qwen 3.6 Flash 17
  235. 235 Virtuoso Large 17
  236. 236 Qwen3-4B 16
  237. 237 Kimi Latest 16
  238. 238 MAI DS R1 16
  239. 239 Seed 1.6 Flash 16
  240. 240 Qwen2.5 7B Instruct 16
  241. 241 WizardLM-2 8x22B 16
  242. 242 GPT-4.1 nano 16
  243. 243 Qwen3-30B-A3B 16
  244. 244 Nano Banana 2 (Gemini 3.1 Flash Image Preview) 16
  245. 245 Relace Apply 3 16
  246. 246 Ring-1T 16
  247. 247 MiniMax M2-her 16
  248. 248 Maestro Reasoning 15
  249. 249 Hunyuan-A13B-Instruct 15
  250. 250 LFM2-8B-A1B 15
  251. 251 Ling-1T 15
  252. 252 Skyfall 36B V2 15
  253. 253 Kimi-Dev-72B 15
  254. 254 GPT-5.4 Mini 15
  255. 255 Mistral NeMo 15
  256. 256 CodeLLaMa 7B Instruct Solidity 15
  257. 257 Kimi Linear 48B A3B Instruct 15
  258. 258 Mistral 7B Instruct v0.3 15
  259. 259 QwQ 32B RpR v1 15
  260. 260 Mistral-7B-Instruct-v0.2 15
  261. 261 Aion-RP 1.0 (8B) 14
  262. 262 Llama 3.2 3B Instruct 14
  263. 263 ERNIE 4.5 VL 28B A3B 14
  264. 264 Jamba Mini 1.7 14
  265. 265 Llama 3 8B Lunaris 14
  266. 266 UI-TARS 7B 14
  267. 267 Llama Guard 3 8B 14
  268. 268 Command R7B 14
  269. 269 Anubis 70B V1.1 13
  270. 270 GPT Audio 13
  271. 271 LFM2-2.6B 13
  272. 272 Mixtral 8x7B Instruct 13
  273. 273 Nova Lite 1.0 13
  274. 274 DeepHermes 3 Mistral 24B Preview 13
  275. 275 AFM-4.5B 13
  276. 276 GPT-4o Audio 13
  277. 277 MiniMax-01 13
  278. 278 MythoMax 13B 13
  279. 279 Qwen3 VL 8B Instruct 13
  280. 280 GPT-5.4 Nano 13
  281. 281 Morph V3 Large 13
  282. 282 GLM-4.1V-9B-Thinking 13
  283. 283 R1 (free) 13
  284. 284 Phi-3-mini-128k-instruct 12
  285. 285 GPT-4o-mini Search Preview 12
  286. 286 Granite 4.1 8B 12
  287. 287 Qianfan-OCR-Fast 12
  288. 288 Hermes 2 Pro - Llama-3 8B 12
  289. 289 Reka Flash 3 11
  290. 290 Llama 3 8B Instruct 11
  291. 291 Olmo 3 7B Instruct 10
  292. 292 ERNIE 4.5 21B A3B 10
  293. 293 Phi-3.5 Mini 128K Instruct 10
  294. 294 Phi 4 Mini Instruct 10
  295. 295 Nova Micro 1.0 9
  296. 296 DeepSeek-V3.2-Speciale 9
  297. 297 Llama 3.2 1B Instruct 9
  298. 298 Ministral 3B 9
  299. 299 DeepSeek V4 Pro 0423 9
  300. 300 Morph V3 Fast 9
  301. 301 Relace Search 9
  302. 302 GPT-4o Mar 2025 9
  303. 303 Qwen2.5-VL-7B 8
  304. 304 GPT Audio Mini 8
  305. 305 Gemma 3 4B 8
  306. 306 Spotlight 8
  307. 307 Reka Edge 4
  308. 308 Ling-2.6-flash 1
  309. 309 granite-4.0-micro 1
#EntryScoreSettingDateSource
1 GPT-5.1-Codex-Max 38 Nov 2025 Official leaderboard
2 Gemini 3 Pro 38 Nov 2025 Official leaderboard
3 GPT-6 Astra 38 Sep 2026 Official leaderboard
4 GPT-5.5 37 Apr 2026 Official leaderboard
5 Gemini 3.1 Pro 37 Feb 2026 Official leaderboard
6 GPT-5.1-Codex-mini 37 Nov 2025 Official leaderboard
7 o3 36 Apr 2025 Official leaderboard
8 Claude Opus 4.8 35 May 2026 Official leaderboard
9 GPT-5 35 Aug 2025 Official leaderboard
10 Gemini 1.0 Pro 35 Dec 2023 Official leaderboard
11 o4-mini 35 Apr 2025 Official leaderboard
12 OpenAI GPT Latest 34 Official leaderboard
13 Qwen3.7-Max 34 May 2026 Official leaderboard
14 Gemini 3.5 Flash 34 May 2026 Official leaderboard
15 GPT-5.3 Codex 33 Feb 2026 Official leaderboard
16 GPT-5-Codex 33 Sep 2025 Official leaderboard
17 Gemini 2.0 Flash Feb 2025 33 Feb 2025 Official leaderboard
18 Grok 4 Fast 33 Sep 2025 Official leaderboard
19 Claude Opus 4.6 33 Feb 2026 Official leaderboard
20 Claude Sonnet 4.5 33 Sep 2025 Official leaderboard
21 GPT-5.1 33 Nov 2025 Official leaderboard
22 GPT-5.2 33 Dec 2025 Official leaderboard
23 Qwen3.5-122B-A10B 33 Feb 2026 Official leaderboard
24 Qwen3 VL 30B A3B Thinking 33 Official leaderboard
25 GPT-5.2 Codex 32 Dec 2025 Official leaderboard
26 Qwen3.5-Flash 32 Feb 2026 Official leaderboard
27 GPT-4.1 32 Apr 2025 Official leaderboard
28 Claude Sonnet 4 32 May 2025 Official leaderboard
29 GPT-5.1-Codex 32 Nov 2025 Official leaderboard
30 Mercury 32 Jun 2025 Official leaderboard
31 Gemini 3 Flash 32 Dec 2025 Official leaderboard
32 Gemini 2.5 Flash Sep 2025 31 Sep 2025 Official leaderboard
33 DeepSeek V3.1 Terminus (exacto) 31 Official leaderboard
34 gpt-oss-safeguard-20b 31 Official leaderboard
35 Gemini 3.1 Pro Preview Custom Tools 31 Official leaderboard
36 Qwen3.5-27B 31 Feb 2026 Official leaderboard
37 Qwen 3.6 Max (Preview) 31 Apr 2026 Official leaderboard
38 Qwen3-Max 31 Sep 2025 Official leaderboard
39 Gemini 3.1 Flash-Lite 31 Mar 2026 Official leaderboard
40 GPT-5 Image 31 Official leaderboard
41 GLM-5 31 Feb 2026 Official leaderboard
42 R1 Distill Llama 70B 30 Official leaderboard
43 Kimi K2 0905 30 Official leaderboard
44 Nano Banana (Gemini 2.5 Flash Image) 30 Official leaderboard
45 Claude Opus 4.5 30 Nov 2025 Official leaderboard
46 Claude Opus 4.7 30 Apr 2026 Official leaderboard
47 GPT-5.6 Terra 30 Jul 2026 Official leaderboard
48 OpenAI deep research 30 Feb 2025 Official leaderboard
49 Qwen3 235B A22B Instruct 2507 29 Official leaderboard
50 Claude Sonnet Latest 29 Official leaderboard
51 Gemini 2.5 Pro Jun 2025 29 Jun 2025 Official leaderboard
52 Kimi K2 Thinking 29 Nov 2025 Official leaderboard
53 GPT-5.4 29 Mar 2026 Official leaderboard
54 GPT-5 Image Mini 29 Official leaderboard
55 Claude Opus Latest 29 Official leaderboard
56 Mercury Coder 29 Official leaderboard
57 Qwen3-VL-235B-A22B-Instruct 29 Sep 2025 Official leaderboard
58 GPT Chat Latest 28 Official leaderboard
59 Codex Mini 28 Official leaderboard
60 Mercury 2 28 Feb 2026 Official leaderboard
61 GLM-5V-Turbo 28 Apr 2026 Official leaderboard
62 DeepSeek V4 Flash 0423 28 Official leaderboard
63 o3-mini 28 Jan 2025 Official leaderboard
64 grok-code-fast-1 28 Aug 2025 Official leaderboard
65 Qwen 3.6 Plus 28 Mar 2026 Official leaderboard
66 GPT-5 nano 28 Aug 2025 Official leaderboard
67 R1 Distill Qwen 14B 28 Official leaderboard
68 Qwen-VL-Max 28 Jan 2024 Official leaderboard
69 Mistral Small 3.2 24B 27 Official leaderboard
70 KAT-Coder-Pro V2 27 Official leaderboard
71 Qwen3-Next-80B-A3B-Instruct 27 Sep 2025 Official leaderboard
72 MiMo-V2.5-Pro 27 Apr 2026 Official leaderboard
73 Claude 3.7 Sonnet 27 thinking Feb 2025 Official leaderboard
74 MiniMax-M2.5 27 Feb 2026 Official leaderboard
75 Qwen3.5 Plus 2026-02-15 27 Official leaderboard
76 Claude Sonnet 4.6 26 Feb 2026 Official leaderboard
77 Gemini 2.5 Flash Image Preview (nano-banana) 26 Official leaderboard
78 Qwen-Turbo 26 Nov 2024 Official leaderboard
79 Qwen3-Next-80B-A3B 26 thinking Sep 2025 Official leaderboard
80 GLM-5-Turbo 26 Mar 2026 Official leaderboard
81 Qwen3 30B A3B Thinking 2507 26 Official leaderboard
82 GPT-5 Chat 26 Aug 2025 Official leaderboard
83 R1 0528 26 Official leaderboard
84 GLM-4.6V 26 Dec 2025 Official leaderboard
85 DeepSeek R1T Chimera 26 Official leaderboard
86 GLM-5.1 26 Apr 2026 Official leaderboard
87 Qwen Plus Apr 2025 26 Apr 2025 Official leaderboard
88 gpt-oss-120b 26 Aug 2025 Official leaderboard
89 GLM-4.7 26 Dec 2025 Official leaderboard
90 Llama-3.1-Nemotron-70B-Instruct 26 Oct 2024 Official leaderboard
91 Magistral Medium 1.0 26 Jun 2025 Official leaderboard
92 Sonar Pro Search 26 Official leaderboard
93 Tongyi DeepResearch 30B A3B 26 Official leaderboard
94 GPT-5 mini 26 Aug 2025 Official leaderboard
95 Claude Haiku 4.5 26 Oct 2025 Official leaderboard
96 GPT-5.2 Instant 26 Dec 2025 Official leaderboard
97 Devstral Small 1.1 25 Jul 2025 Official leaderboard
98 ERNIE 4.5 VL 424B A47B 25 Official leaderboard
99 GPT-4o-mini (2024-07-18) 25 Official leaderboard
100 Olmo 3 7B Think 25 Official leaderboard
101 GLM-4.6 25 Sep 2025 Official leaderboard
102 R1T Chimera 25 Official leaderboard
103 Gemini 2.5 Flash-Lite Sep 2025 25 Sep 2025 Official leaderboard
104 Qwen3.5-35B-A3B 25 Feb 2026 Official leaderboard
105 Qwen3-8B 25 Apr 2025 Official leaderboard
106 Llama 3.2 90B 25 Sep 2024 Official leaderboard
107 Mistral Small 3 25 Jan 2025 Official leaderboard
108 GPT-4.1 mini 25 Apr 2025 Official leaderboard
109 Doubao-Seed-1.6 25 Jun 2025 Official leaderboard
110 qwen3-coder-plus 25 Sep 2025 Official leaderboard
111 Qwen3-VL-235B-A22B 25 thinking Sep 2025 Official leaderboard
112 Ling-2.6-1T 25 Apr 2026 Official leaderboard
113 Claude Opus 4.8 (Fast) 25 Official leaderboard
114 MiniMax-M3 25 Jun 2026 Official leaderboard
115 Devstral 2 2512 25 Official leaderboard
116 QwQ 32B RpR v1 (free) 25 Official leaderboard
117 Gemma 3 27B 25 Mar 2025 Official leaderboard
118 Kimi K2.5 25 Jan 2026 Official leaderboard
119 GPT-5.6 Sol 25 Jul 2026 Official leaderboard
120 DeepSeek V3 0324 (free) 24 Official leaderboard
121 Nemotron 3 Ultra 24 Jun 2026 Official leaderboard
122 MiniMax-M1-80k 24 Jun 2025 Official leaderboard
123 MiniMax-M2 24 Oct 2025 Official leaderboard
124 Qwen 3.6 35B-A3B 24 Apr 2026 Official leaderboard
125 DeepSeek V3.1 Nex N1 24 Official leaderboard
126 Qwen3-32B 24 Apr 2025 Official leaderboard
127 Tencent Hy3 preview 24 Apr 2026 Official leaderboard
128 DeepSeek R1 0528 Qwen3 8B (free) 24 Official leaderboard
129 GLM 4 32B 24 Official leaderboard
130 GLM-4.5-Air 24 Jul 2025 Official leaderboard
131 o4 Mini Deep Research 24 Official leaderboard
132 Gemma 4 26B A4B 24 Apr 2026 Official leaderboard
133 gpt-oss-120b (exacto) 24 Official leaderboard
134 Seed2.0 Lite (0215) 24 Feb 2026 Official leaderboard
135 Claude Haiku Latest 23 Official leaderboard
136 DeepSeek-V3.1 23 Aug 2025 Official leaderboard
137 Step 3.5 Flash 23 Feb 2026 Official leaderboard
138 Qwen3.7-Plus 23 Jun 2026 Official leaderboard
139 GPT-5.1 Chat 23 Official leaderboard
140 Aion-1.0 23 Official leaderboard
141 GPT-4o Search Preview 23 Official leaderboard
142 ERNIE-4.5-300B-A47B 23 Jun 2025 Official leaderboard
143 Cydonia 24B V4.1 23 Official leaderboard
144 Llama 4 Maverick 23 Apr 2025 Official leaderboard
145 Aion-1.0-Mini 23 Official leaderboard
146 Seed2.0 Mini 23 Feb 2026 Official leaderboard
147 Qwen3-Coder-30B-A3B-Instruct 23 Jul 2025 Official leaderboard
148 Ring-2.6-1T 23 May 2026 Official leaderboard
149 Qwen Plus 0728 (thinking) 23 Official leaderboard
150 MiMo-V2-Flash 23 Dec 2025 Official leaderboard
151 Qwen3.5 397B-A17B 23 Feb 2026 Official leaderboard
152 Aion-2.0 22 Official leaderboard
153 MiniMax-M2.7 22 Mar 2026 Official leaderboard
154 Olmo 3.1 32B Think 22 Dec 2025 Official leaderboard
155 Mistral Small 3.1 24B 21 Official leaderboard
156 Mistral-7B-Instruct-v0.1 21 Sep 2023 Official leaderboard
157 Qwen3 30B A3B Instruct 2507 21 Official leaderboard
158 Llama 3.3 70B Instruct 21 Dec 2024 Official leaderboard
159 Gemini 2.0 Flash-Lite 21 Feb 2025 Official leaderboard
160 Qwen3-Coder-480B-A35B 21 Jul 2025 Official leaderboard
161 MiMo-V2.5 21 Apr 2026 Official leaderboard
162 GPT-5.3 Instant 21 Mar 2026 Official leaderboard
163 Step 3.7 Flash 21 May 2026 Official leaderboard
164 Llama 3.3 Euryale 70B 21 Official leaderboard
165 Pixtral Large 2411 21 Official leaderboard
166 Palmyra X5 21 Apr 2025 Official leaderboard
167 GLM-4.5V 21 Aug 2025 Official leaderboard
168 DeepSeek-V3.1-Terminus 21 Sep 2025 Official leaderboard
169 Hermes 3 405B Instruct 21 Official leaderboard
170 Llama 4 Scout 21 Apr 2025 Official leaderboard
171 qwen3-coder-next 21 Feb 2026 Official leaderboard
172 MiMo-V2-Pro 21 Mar 2026 Official leaderboard
173 GPT-5.6 Luna 21 Jul 2026 Official leaderboard
174 DeepSeek-V3 Mar 2025 21 Mar 2025 Official leaderboard
175 Hermes 4 405B 21 Aug 2025 Official leaderboard
176 Qwen3-VL-8B-Thinking 21 Oct 2025 Official leaderboard
177 Qwen Plus 0728 20 Official leaderboard
178 Qwen3-235B-A22B 20 Apr 2025 Official leaderboard
179 MiMo-V2-Omni 20 Mar 2026 Official leaderboard
180 Hermes 4 70B 20 Official leaderboard
181 Llama 3 70B Instruct 20 Official leaderboard
182 Perceptron Mk1 20 Official leaderboard
183 Qwen-Max 20 Official leaderboard
184 Phi-4-multimodal 20 Feb 2025 Official leaderboard
185 Gemini Flash Latest 20 Official leaderboard
186 LongCat-Flash-Chat 20 Aug 2025 Official leaderboard
187 DeepSeek-V3.2 20 Dec 2025 Official leaderboard
188 GPT Mini Latest 20 Official leaderboard
189 Qwen2.5-Coder-32B-Instruct 20 Nov 2024 Official leaderboard
190 Qwen2.5-VL-72B 20 Jan 2025 Official leaderboard
191 Qwen3-14B 20 Apr 2025 Official leaderboard
192 MiniMax-M2.1 20 Dec 2025 Official leaderboard
193 GPT-4o mini 20 Jul 2024 Official leaderboard
194 Trinity Large Thinking 20 Apr 2026 Official leaderboard
195 InternVL3-78B 19 Apr 2025 Official leaderboard
196 Olmo 3 32B Think 19 Nov 2025 Official leaderboard
197 Gemma 2 9B 19 Jun 2024 Official leaderboard
198 Gemma 3 12B 19 Mar 2025 Official leaderboard
199 Magistral Small 2506 19 Official leaderboard
200 Sonar Pro 19 Jan 2025 Official leaderboard
201 Ministral 8B 19 Oct 2024 Official leaderboard
202 LlamaGuard 2 8B 19 Official leaderboard
203 Gemma 3n 4B 19 Official leaderboard
204 Qwen3 Coder 480B A35B (exacto) 18 Official leaderboard
205 ERNIE-4.5-21B-A3B-Thinking 18 Sep 2025 Official leaderboard
206 Qwen3.6 27B 18 Apr 2026 Official leaderboard
207 INTELLECT-3 18 Nov 2025 Official leaderboard
208 Llama 3.1 8B Instruct 18 Official leaderboard
209 Llama Guard 4 12B 18 Official leaderboard
210 Qwen3 Coder Flash 18 Official leaderboard
211 OLMo 2 32B Instruct 18 Mar 2025 Official leaderboard
212 KAT-Coder-Pro V1 18 Official leaderboard
213 Nemotron Nano 9B V2 18 Official leaderboard
214 Llama 3.2 11B 18 Sep 2024 Official leaderboard
215 Command A 18 Official leaderboard
216 Voxtral Small 18 Jul 2025 Official leaderboard
217 glm-4.7-flash 18 Jan 2026 Official leaderboard
218 Gemma 4 31B IT 18 Apr 2026 Official leaderboard
219 Jamba Large 1.7 18 Official leaderboard
220 Qwen2.5-72B 18 Sep 2024 Official leaderboard
221 Phi-4 18 Dec 2024 Official leaderboard
222 Step 3 18 Jul 2025 Official leaderboard
223 Llama 3 Euryale 70B v2.1 18 Official leaderboard
224 Qwen2.5 Coder 7B Instruct 18 Official leaderboard
225 Qwen3.5 Plus 2026-04-20 17 Official leaderboard
226 Pixtral 12B 17 Sep 2024 Official leaderboard
227 Nova Premier 1.0 17 Official leaderboard
228 deepseek-r1-0528-qwen3-8b 17 May 2025 Official leaderboard
229 Qwen2.5-VL-32B 17 Mar 2025 Official leaderboard
230 GLM-4.5 17 Aug 2025 Official leaderboard
231 DeepSeek R1T2 Chimera 17 Official leaderboard
232 GPT-5.4 Image 2 17 Official leaderboard
233 Qwen3-235B-A22B-Thinking Jul 2025 17 Jul 2025 Official leaderboard
234 Qwen 3.6 Flash 17 Apr 2026 Official leaderboard
235 Virtuoso Large 17 Official leaderboard
236 Qwen3-4B 16 Apr 2025 Official leaderboard
237 Kimi Latest 16 Official leaderboard
238 MAI DS R1 16 Official leaderboard
239 Seed 1.6 Flash 16 Official leaderboard
240 Qwen2.5 7B Instruct 16 Official leaderboard
241 WizardLM-2 8x22B 16 Apr 2024 Official leaderboard
242 GPT-4.1 nano 16 Apr 2025 Official leaderboard
243 Qwen3-30B-A3B 16 Apr 2025 Official leaderboard
244 Nano Banana 2 (Gemini 3.1 Flash Image Preview) 16 Official leaderboard
245 Relace Apply 3 16 Official leaderboard
246 Ring-1T 16 Oct 2025 Official leaderboard
247 MiniMax M2-her 16 Official leaderboard
248 Maestro Reasoning 15 Official leaderboard
249 Hunyuan-A13B-Instruct 15 Jun 2025 Official leaderboard
250 LFM2-8B-A1B 15 Oct 2025 Official leaderboard
251 Ling-1T 15 Official leaderboard
252 Skyfall 36B V2 15 Official leaderboard
253 Kimi-Dev-72B 15 Jun 2025 Official leaderboard
254 GPT-5.4 Mini 15 Mar 2026 Official leaderboard
255 Mistral NeMo 15 Jul 2024 Official leaderboard
256 CodeLLaMa 7B Instruct Solidity 15 Official leaderboard
257 Kimi Linear 48B A3B Instruct 15 Official leaderboard
258 Mistral 7B Instruct v0.3 15 Official leaderboard
259 QwQ 32B RpR v1 15 Official leaderboard
260 Mistral-7B-Instruct-v0.2 15 Dec 2023 Official leaderboard
261 Aion-RP 1.0 (8B) 14 Official leaderboard
262 Llama 3.2 3B Instruct 14 Official leaderboard
263 ERNIE 4.5 VL 28B A3B 14 Official leaderboard
264 Jamba Mini 1.7 14 Official leaderboard
265 Llama 3 8B Lunaris 14 Official leaderboard
266 UI-TARS 7B 14 Official leaderboard
267 Llama Guard 3 8B 14 Official leaderboard
268 Command R7B 14 Dec 2024 Official leaderboard
269 Anubis 70B V1.1 13 Official leaderboard
270 GPT Audio 13 Official leaderboard
271 LFM2-2.6B 13 Official leaderboard
272 Mixtral 8x7B Instruct 13 Official leaderboard
273 Nova Lite 1.0 13 Official leaderboard
274 DeepHermes 3 Mistral 24B Preview 13 Official leaderboard
275 AFM-4.5B 13 Jul 2025 Official leaderboard
276 GPT-4o Audio 13 Official leaderboard
277 MiniMax-01 13 Official leaderboard
278 MythoMax 13B 13 Official leaderboard
279 Qwen3 VL 8B Instruct 13 Official leaderboard
280 GPT-5.4 Nano 13 Mar 2026 Official leaderboard
281 Morph V3 Large 13 Official leaderboard
282 GLM-4.1V-9B-Thinking 13 Jul 2025 Official leaderboard
283 R1 (free) 13 Official leaderboard
284 Phi-3-mini-128k-instruct 12 Apr 2024 Official leaderboard
285 GPT-4o-mini Search Preview 12 Official leaderboard
286 Granite 4.1 8B 12 Apr 2026 Official leaderboard
287 Qianfan-OCR-Fast 12 Official leaderboard
288 Hermes 2 Pro - Llama-3 8B 12 Official leaderboard
289 Reka Flash 3 11 Mar 2025 Official leaderboard
290 Llama 3 8B Instruct 11 Official leaderboard
291 Olmo 3 7B Instruct 10 Official leaderboard
292 ERNIE 4.5 21B A3B 10 Official leaderboard
293 Phi-3.5 Mini 128K Instruct 10 Official leaderboard
294 Phi 4 Mini Instruct 10 Official leaderboard
295 Nova Micro 1.0 9 Official leaderboard
296 DeepSeek-V3.2-Speciale 9 Dec 2025 Official leaderboard
297 Llama 3.2 1B Instruct 9 Official leaderboard
298 Ministral 3B 9 Oct 2024 Official leaderboard
299 DeepSeek V4 Pro 0423 9 Official leaderboard
300 Morph V3 Fast 9 Official leaderboard
301 Relace Search 9 Official leaderboard
302 GPT-4o Mar 2025 9 Mar 2025 Official leaderboard
303 Qwen2.5-VL-7B 8 Jan 2025 Official leaderboard
304 GPT Audio Mini 8 Official leaderboard
305 Gemma 3 4B 8 Mar 2025 Official leaderboard
306 Spotlight 8 Official leaderboard
307 Reka Edge 4 Mar 2026 Official leaderboard
308 Ling-2.6-flash 1 Official leaderboard
309 granite-4.0-micro 1 Oct 2025 Official leaderboard

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
o4-mini
high
35
Claude 3.7 Sonnet
thinking
27
Magistral Medium 1.0
thinking
26
ModelEffortHarnessScoreCost per taskOutput tokens
GPT-5.1-Codex-Max 38
Gemini 3 Pro 38
GPT-6 Astra 38
GPT-5.5 37
Gemini 3.1 Pro 37
GPT-5.1-Codex-mini 37
o3 36
Claude Opus 4.8 35
GPT-5 35
Gemini 1.0 Pro 35
o4-mini 35
Qwen3.7-Max 34
Gemini 3.5 Flash 34
o4-mini high 34
GPT-5.3 Codex 33
GPT-5-Codex 33
Gemini 2.0 Flash Feb 2025 33
Grok 4 Fast 33
Claude Opus 4.6 33
Claude Sonnet 4.5 33
GPT-5.1 33
GPT-5.2 33
Qwen3.5-122B-A10B 33
GPT-5.2 Codex 32
Qwen3.5-Flash 32
GPT-4.1 32
Claude Sonnet 4 32
GPT-5.1-Codex 32
Mercury 32
Gemini 3 Flash 32
Gemini 2.5 Flash Sep 2025 31
Qwen3.5-27B 31
Qwen 3.6 Max (Preview) 31
Qwen3-Max 31
Gemini 2.5 Flash Sep 2025 31
Gemini 3.1 Flash-Lite 31
GLM-5 31
Claude Opus 4.5 30
Claude Opus 4.7 30
GPT-5.6 Terra 30
OpenAI deep research 30
Gemini 2.5 Pro Jun 2025 29
Kimi K2 Thinking 29
GPT-5.4 29
Qwen3-VL-235B-A22B-Instruct 29
Gemini 2.5 Pro Jun 2025 28
Gemini 3.1 Flash-Lite 28
Mercury 2 28
GLM-5V-Turbo 28
o3-mini 28
grok-code-fast-1 28
Qwen 3.6 Plus 28
GPT-5 nano 28
Qwen-VL-Max 28
Gemini 2.5 Pro Jun 2025 28
Qwen3-Next-80B-A3B-Instruct 27
MiMo-V2.5-Pro 27
Claude 3.7 Sonnet thinking 27
MiniMax-M2.5 27
Claude Sonnet 4.6 26
Gemini 2.5 Flash Image Preview (nano-banana) 26
Qwen-Turbo 26
Qwen3-Next-80B-A3B thinking 26
GLM-5-Turbo 26
GPT-5 Chat 26
GLM-4.6V 26
o3-mini 26
GLM-5.1 26
Qwen Plus Apr 2025 26
gpt-oss-120b 26
GLM-4.7 26
Llama-3.1-Nemotron-70B-Instruct 26
Magistral Medium 1.0 26
GPT-5 mini 26
Claude Haiku 4.5 26
GPT-5.2 Instant 26
Devstral Small 1.1 25
GLM-4.6 25
Gemini 2.5 Flash-Lite Sep 2025 25
Qwen3.5-35B-A3B 25
Claude 3.7 Sonnet 25
Qwen3-8B 25
Llama 3.2 90B 25
Mistral Small 3 25
GPT-4.1 mini 25
Doubao-Seed-1.6 25
qwen3-coder-plus 25
Qwen3-VL-235B-A22B thinking 25
Ling-2.6-1T 25
MiniMax-M3 25
Gemma 3 27B 25
Kimi K2.5 25
GPT-5.6 Sol 25
Nemotron 3 Ultra 24
MiniMax-M1-80k 24
MiniMax-M2 24
Qwen 3.6 35B-A3B 24
Qwen3-32B 24
Tencent Hy3 preview 24
GLM-4.5-Air 24
Gemma 4 26B A4B 24
Seed2.0 Lite (0215) 24
DeepSeek-V3.1 23
Step 3.5 Flash 23
Qwen3.7-Plus 23
Gemini 2.5 Flash-Lite Sep 2025 23
ERNIE-4.5-300B-A47B 23
Llama 4 Maverick 23
Seed2.0 Mini 23
Qwen3-Coder-30B-A3B-Instruct 23
Ring-2.6-1T 23
MiMo-V2-Flash 23
Qwen3.5 397B-A17B 23
MiniMax-M2.7 22
Olmo 3.1 32B Think 22
Mistral-7B-Instruct-v0.1 21
Llama 3.3 70B Instruct 21
Gemini 2.0 Flash-Lite 21
Qwen3-Coder-480B-A35B 21
MiMo-V2.5 21
GPT-5.3 Instant 21
Step 3.7 Flash 21
Palmyra X5 21
GLM-4.5V 21
DeepSeek-V3.1-Terminus 21
Llama 4 Scout 21
Magistral Medium 1.0 thinking 21
qwen3-coder-next 21
MiMo-V2-Pro 21
GPT-5.6 Luna 21
DeepSeek-V3 Mar 2025 21
Hermes 4 405B 21
Qwen3-VL-8B-Thinking 21
Qwen3-235B-A22B 20
MiMo-V2-Omni 20
Phi-4-multimodal 20
LongCat-Flash-Chat 20
DeepSeek-V3.2 20
Qwen2.5-Coder-32B-Instruct 20
Qwen2.5-VL-72B 20
Qwen3-14B 20
MiniMax-M2.1 20
GPT-4o mini 20
Trinity Large Thinking 20
InternVL3-78B 19
Olmo 3 32B Think 19
Gemma 2 9B 19
Gemma 3 12B 19
Sonar Pro 19
Ministral 8B 19
ERNIE-4.5-21B-A3B-Thinking 18
Qwen3.6 27B 18
INTELLECT-3 18
OLMo 2 32B Instruct 18
Llama 3.2 11B 18
Voxtral Small 18
glm-4.7-flash 18
Gemma 4 31B IT 18
Qwen2.5-72B 18
Phi-4 18
Step 3 18
Qwen3-Max 18
Pixtral 12B 17
deepseek-r1-0528-qwen3-8b 17
DeepSeek-V3 Mar 2025 17
Qwen2.5-VL-32B 17
GLM-4.5 17
Qwen3-235B-A22B-Thinking Jul 2025 17
Qwen 3.6 Flash 17
Qwen3-4B 16
WizardLM-2 8x22B 16
GPT-4.1 nano 16
Qwen3-30B-A3B 16
Gemini 2.5 Flash-Lite Sep 2025 16
Ring-1T 16
Hunyuan-A13B-Instruct 15
LFM2-8B-A1B 15
Kimi-Dev-72B 15
GPT-5.4 Mini 15
Mistral NeMo 15
Mistral-7B-Instruct-v0.2 15
Command R7B 14
AFM-4.5B 13
GPT-5.4 Nano 13
GLM-4.1V-9B-Thinking 13
Phi-3-mini-128k-instruct 12
Granite 4.1 8B 12
Reka Flash 3 11
DeepSeek-V3.2-Speciale 9
Ministral 3B 9
GPT-4o Mar 2025 9
Qwen2.5-VL-7B 8
Gemma 3 4B 8
Mistral-7B-Instruct-v0.1 8
Reka Edge 4
granite-4.0-micro 1

5 models have held number one since its release in Feb 2025

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

Google DeepMind OpenAI
Own leaderboard Not tested since Sep 2026

Scores over time

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

On the benchmark’s own leaderboard
0 10 20 30 40 2024 2025 2026 0 10 20 30 40 2024 2025 2026

No lab model card reports it yet

We read the headline results of 479 launch posts from 51 labs, and none of them list this benchmark.

More in Games and long tasks

All 68 in Games and long tasks
  • ARC-AGI-3 In use Interactive game environments where agents must explore, infer goals and rules, and act efficiently without instructions. GPT-6 Astra 100%
  • MineDojo Fading Agents complete open-ended tasks in Minecraft, such as mining blocks and fighting mobs, from a large simulation suite. Programmatic checks score many tasks. Seed2.0 Pro 49%
  • Orak Fading LLM agents play 12 popular video games across most genres through an MCP interface. Scores come from in-game progress or outcomes per title. Nemotron 3 Nano 4B 22.9
  • GameWorld Only in system cards Multimodal game agents play 34 browser games across 170 tasks. Outcomes are checked from the game state. Gemini 3 Flash 41.9%
  • Swarm Bench Only in system cards Models orchestrate agent swarms on complex tasks that benefit from coordinated decomposition and parallel execution. Kimi K3 76.3%
  • TextQuests Only in system cards Agents play 25 classic interactive fiction games. Measures game progress and harmful actions per 100 steps. Muse Spark 2.40