Benchmark Almanac

Decision Index

KnowLine 31B Gen1 71.9 tops its leaderboard Gemma 4 31B IT 67.6 leads the models we track Not in model cards

Decision models read a state and a typed question and return a decision. The index weights private same-skill sets and new-domain sets, equated on 21 base models.

Tracks reproductions of TypeSafe's Jev decision model. Version 0.4.1 board lists 248 entries. A separate vision board exists.

Made by Hugging Face community (apolinario, multimodalart) Released Sep 2026 Composite indices Public tasks plus a private test set Last model added Aug 2026 Scores refreshed 2 Nov 2026
Level
Everyday Most adults, without training
Scored
Checked automatically A program checks the result against an answer key, unit tests or the expected final state.
Answers
Prediction Probabilities for events that have not happened yet.
244
Entries on its own leaderboard, 3 models we track

KnowLine 31B Gen1 tops the leaderboard with 71.9

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. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is Gemma 4 31B IT, at 67.6.

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

244 entries. Scroll sideways for the rest.

  1. 71.9 KnowLine 31B Gen1
  2. 71.5 Fastino GLiDE 1.2
  3. 71.1 Sieve 31B v3
  4. 70.5 Turion Jev 31B
  5. 70.2 vLLM-SR d3-ultra (Decision 3.0, 27B)
  6. 69.8 Jeffan-31b-1710
  7. 68.6 H2O Lightning 31B v1.1
  8. 68.5 WeMM-JEV-27B
  9. 68.4 baselabs mammoth-1 (27B)
  10. 68.2 Ainvest-Decision-27B v1.0
  11. 68.1 oenogarum (Gemma 4 31B)
  12. 68.0 Perplexity Decider v1.2 (27B)
  13. 68.0 AutoTrust-JEV-27B v1.1
  14. 67.9 StartLux-Decision v1.1 27B
  15. 67.6 MISA AI 2.0 Decision Plus · oak 1.1 (31B)
  16. 67.6 AJev lm6q (Gemma 4 31B)
  17. 67.6 Gemma 4 31B IT
  18. 67.3 Lychee-27B
  19. 67.3 vLLM-SR d3-turbo v1.1 (27B)
  20. 67.2 vLLM-SR d3 (27B)
  21. 67.0 allec-oavg (Gemma 4 31B)
  22. 66.9 VisionJEV 1.0.0 (27B)
  23. 66.7 Aleria Judge 1.1.1 (Gemma 4 31B)
  24. 66.6 Nace.AI Drex Max v1 (27B)
  25. 66.6 Bespoke Nimble v4 Max (27B)
  26. 66.5 AnyJev Tacitus 31B
  27. 66.0 Decider 31B (NVFP4)
  28. 65.8 KnowLine 35B-A3B Gen1.5
  29. 65.7 OneShot 27B
  30. 65.3 vera-large
  31. 64.8 Bobcat 1.3 (31B)
  32. 64.8 Amos 1 Large (27B)
  33. 64.8 Matilda Decision v1
  34. 64.6 KnowLine 26B-A4B Gen2.5
  35. 64.4 frontier-infra Jebadiah 27B v2.1 (t1e)
  36. 63.0 deck31b (Gemma 4 31B)
  37. 63.0 Darwin-27B-ZTC
  38. 62.7 decisio 0.6.0 (Gemma 4 31B)
  39. 61.7 Cerno 27B
  40. 61.5 Torchcast Decision 27B
  41. 61.5 OneJev1.6-27B
  42. 61.1 Qwen Decision 27B noncausal LoRA
  43. 60.8 Quyet-1.0-Large
  44. 60.3 garum (Gemma 4 31B)
  45. 60.3 baselabs elephant-1 (9B)
  46. 60.1 Standard One 35B SH (v1)
  47. 60.0 Lumma-fev-27B
  48. 59.9 Sieve 27B v2
  49. 59.8 mogan-decision-31B-it
  50. 59.6 linne-fern 1.0 (27B)
  51. 59.6 Kev 27B
  52. 59.4 Xor 26B-A4B v1.2
  53. 59.2 Jev 1.13
  54. 59.1 span-02 (Gemma 4 26B-A4B)
  55. 59.0 garas (Gemma 4 31B)
  56. 58.8 Shisa DE-2.1 (repeat)
  57. 58.8 Eikos-27B
  58. 58.8 reflex Qwen3.8-27B-FP8 (pmscr)
  59. 58.6 RSI-Jev v6.1 VL 27B (effort auto)
  60. 58.4 KnowLine 9B Gen1
  61. 58.1 StartLux-Decision v1.1 9B
  62. 58.0 vLLM-SR Decision 2.0 Vega 27B
  63. 57.6 JEV-35B
  64. 57.4 JEV-27B
  65. 57.0 Qwen 3.8 27B
  66. 56.6 Tern-35B-A3B
  67. 56.4 Surogate Rune 26B-A4B v3
  68. 56.2 SynACK Decide 26B-A4B v7
  69. 56.0 KnowLine 4B Gen5.5
  70. 55.9 Blink v0.3 26B-A4B (NVFP4)
  71. 55.6 JADE
  72. 55.4 Wald-26B-A4B
  73. 55.3 AJev lora1 (Gemma 4 26B-A4B)
  74. 54.9 Cloudflare clef
  75. 54.8 StartLux-Decision v1.1 4B
  76. 54.7 Qwen3.6 27B
  77. 54.3 Perplexity Decider Flash v1.2 (4B)
  78. 54.3 vLLM-SR d3-flash (9B)
  79. 54.1 JPT-35B-A3B
  80. 53.6 Xor 1.2
  81. 53.5 GEV-26B-Decide
  82. 53.0 OneJev1.6-9B
  83. 52.8 Nace.AI Drex v2.0 (35B-A3B)
  84. 52.5 muria (Gemma 4 12B)
  85. 52.2 Jeffan-4B-2010
  86. 52.0 Verdin-4B v1.1
  87. 51.1 Bespoke Nimble 9B v3
  88. 51.0 Senova-4B v2
  89. 50.9 frontier-infra Jebadiah 9B v2.1 (t1e)
  90. 50.8 decisio 0.4.0 (Qwen3.6-35B-A3B)
  91. 50.0 vLLM-SR d3-mini (4B)
  92. 49.5 JoshuaSP diffusiongemma (open-jev)
  93. 48.9 AJev lora7 (Gemma 4 12B)
  94. 48.9 OneJev1.6-4B
  95. 48.8 decisio 0.4.0 (Gemma 4 12B)
  96. 47.7 Mica v0.5 4B
  97. 47.5 PurpleMIST-Flash-1.0
  98. 47.3 Cerno 9B v1.1
  99. 47.0 Sieve 9B v2
  100. 46.6 Decider 35B-A3B
  101. 46.6 Winnow-12B
  102. 46.3 Seb-9B v2
  103. 46.0 vLLM-SR Decision 2.0 Lux 9B
  104. 45.8 Nace.AI Drex v1.5
  105. 45.8 Hopper 12B (trained)
  106. 45.6 Trio-Spark DI (9B)
  107. 45.5 CoCo-Decision-4B 2.4
  108. 45.0 Cloudflare clef-omni (30B-A3B)
  109. 44.9 JevK5 9B v0.3.3
  110. 44.4 Cloudflare clef-flash
  111. 44.4 Jevfire
  112. 44.3 RSI-Jev v6.1 VL 4B
  113. 44.2 Jegue 0.1 (9B)
  114. 44.1 jiwo 4B v2
  115. 44.1 JPT-9B
  116. 43.9 ezjev 4B s2
  117. 43.8 Jev-Omni
  118. 43.2 Solomon v1.1
  119. 42.9 Standard One 8B SH (v4)
  120. 42.7 Amber Hybrid 4B
  121. 42.5 Saif 4B
  122. 42.1 JEV-9B
  123. 42.1 Bespoke Nimble 9B v2
  124. 41.7 Sieve 9B
  125. 41.1 Cerno 4B v1.1
  126. 41.0 Sieve 4B
  127. 40.9 Kev 9B v2
  128. 40.9 Nari Decision 4B v3.2
  129. 40.8 Aplomb 1
  130. 40.8 vLLM-SR Decision 2.0 Nox 4B
  131. 40.2 Hopper (G) 1.2
  132. 40.1 StartLux-Decision v1.1 2B
  133. 39.8 FlyMy.AI Decision 4B v1.6
  134. 39.8 Standard One 3B SH (v4)
  135. 39.6 JPT-4B
  136. 39.5 Amber 4B
  137. 38.7 Intern-Decision-4B
  138. 38.6 RSI-Jev v5.0 VL 3B
  139. 38.1 djev
  140. 37.9 Valen-4B
  141. 37.6 razorback16 openjev diffusiongemma (NVFP4, vLLM)
  142. 37.3 NeoHorse-Jev-4B
  143. 36.6 JevK5 v0.3
  144. 36.3 Decider 4B
  145. 35.9 baselabs joey-1 (2B)
  146. 35.5 lev
  147. 35.4 Lev1 3.7B
  148. 34.9 Kev 4B r10
  149. 34.5 Jev-Style 2B v4
  150. 34.5 Sieve 2B v3
  151. 34.4 LatentMesh Blitz 2B v1
  152. 34.4 Saracura PT-BR 4B
  153. 34.4 Cam1
  154. 34.3 vLLM-SR d3-nano (2B)
  155. 34.3 mmastrac diffusiongemma (vLLM PR 57250)
  156. 34.0 Kas 4B
  157. 33.7 Tev1-4B-experimental
  158. 32.7 Winnow-E4B
  159. 32.5 Aurelis-2B
  160. 32.2 Jevstral 8B
  161. 32.2 Metask-Jev-4B
  162. 32.0 Ryotide Qwen v0.3.1
  163. 31.3 Intelif Qwen3-4B
  164. 31.2 GLiClass v4 edge (0.4B)
  165. 31.1 LiquidAI d1-3B
  166. 31.1 open-jev (pngwn)
  167. 30.6 vera-core-2
  168. 30.4 Jobe
  169. 29.6 OneJev1.6-0.8B
  170. 29.4 crystal-0.8b
  171. 29.3 vLLM-SR vela3-fast (Vela 3.0, 0.4B)
  172. 29.3 Crivo-1B
  173. 29.2 Gevva E4B
  174. 28.9 StartLux-Decision v1.1 0.8B
  175. 28.8 openvons
  176. 26.7 SemIf
  177. 26.5 Saina Helm 2 Omni (0.8B)
  178. 26.2 vLLM-SR d3-lite (0.8B)
  179. 26.0 Amber 0.4B
  180. 25.8 jiwo 0.8B v2
  181. 25.3 Saina Helm 2.0 v2
  182. 24.9 this-that 1.2
  183. 24.3 Gevva E2B
  184. 23.7 Crivo-400M
  185. 23.6 Decider 2B
  186. 23.6 vLLM-SR Decision 2.0 Sol 2B
  187. 23.6 jdi-ptr14-0.8b
  188. 22.8 mini-jev
  189. 22.2 Valen-2B
  190. 22.1 vLLM-SR vela3-fast-vision (Vela 3.0, 0.25B)
  191. 21.9 Sieve 2B
  192. 21.5 wyat-1-800m
  193. 20.9 Intern-Decision-2B
  194. 20.8 Sieve 0.8B
  195. 20.7 Strands Decider 2B
  196. 20.6 EXAONE-4.0-1.2B-JEV v0.3
  197. 19.9 tasksource Decider nano v5 (150M)
  198. 19.5 Crivo-68M
  199. 18.5 JEV Ettin 150M
  200. 18.4 vLLM-SR Decision 2.0 Eos 0.8B
  201. 18.3 Crivo-150M v2
  202. 18.0 Bekko System One v0 400M
  203. 17.8 Mola-150M
  204. 17.7 vLLM-SR d3-edge (0.6B)
  205. 17.6 Mola-68M
  206. 17.3 JEV FLAN-T5 Large public v2 (342M)
  207. 17.2 DecisionTune 1.0
  208. 16.6 Valen-0.8B
  209. 16.5 Bosun v3.1 1.7B
  210. 16.4 vLLM-SR Decision 2.0 Kai 0.6B
  211. 16.4 Dinah-1
  212. 16.3 Kev 0.8B r15
  213. 15.6 PICK-150M v5
  214. 15.5 JPT-0.8B
  215. 15.2 JEV Ettin 400M
  216. 14.5 Sifr 0.8B v3.1
  217. 14.2 Crivo-32M v2
  218. 14.0 Prumo
  219. 13.9 Kodiak v0.2 1B (accuracy)
  220. 12.7 Intern-Decision-0.8B
  221. 12.5 Jev-Style 0.8B v3
  222. 12.5 Tev1-0.8B-experimental
  223. 12.4 Bekko System One v0 68M
  224. 11.8 LiquidAI d1-omni-600M
  225. 11.7 Crivo-17M
  226. 11.7 Von 1.3
  227. 11.0 Bosun v3.1 0.6B
  228. 11.0 vera-spark
  229. 10.8 Lavoir
  230. 10.1 GLiNER2.5-Decide
  231. 9.2 Laya
  232. 8.8 JevK5 Lite
  233. 8.2 LFM2.5-2.6B-RLCD
  234. 7.8 Qwen-2.5-1B-RLCD
  235. 7.7 Bekko System One v0 17M
  236. 6.5 CLM-v0.1-8B
  237. 6.2 jeff
  238. 5.9 LFM2.5-350M-RLCD
  239. 5.2 system-one-gemma
  240. 5.0 DM-JEPA 1.1
  241. 4.8 Julia 1
  242. 3.5 Nirnaya 0.4B
  243. 3.4 Zev
  244. 2.0 Verdict
  1. 1 KnowLine 31B Gen1 71.9
  2. 2 Fastino GLiDE 1.2 71.5
  3. 3 Sieve 31B v3 71.1
  4. 4 Turion Jev 31B 70.5
  5. 5 vLLM-SR d3-ultra (Decision 3.0, 27B) 70.2
  6. 6 Jeffan-31b-1710 69.8
  7. 7 H2O Lightning 31B v1.1 68.6
  8. 8 WeMM-JEV-27B 68.5
  9. 9 baselabs mammoth-1 (27B) 68.4
  10. 10 Ainvest-Decision-27B v1.0 68.2
  11. 11 oenogarum (Gemma 4 31B) 68.1
  12. 12 Perplexity Decider v1.2 (27B) 68.0
  13. 13 AutoTrust-JEV-27B v1.1 68.0
  14. 14 StartLux-Decision v1.1 27B 67.9
  15. 15 MISA AI 2.0 Decision Plus · oak 1.1 (31B) 67.6
  16. 16 AJev lm6q (Gemma 4 31B) 67.6
  17. 17 Gemma 4 31B IT 67.6
  18. 18 Lychee-27B 67.3
  19. 19 vLLM-SR d3-turbo v1.1 (27B) 67.3
  20. 20 vLLM-SR d3 (27B) 67.2
  21. 21 allec-oavg (Gemma 4 31B) 67.0
  22. 22 VisionJEV 1.0.0 (27B) 66.9
  23. 23 Aleria Judge 1.1.1 (Gemma 4 31B) 66.7
  24. 24 Nace.AI Drex Max v1 (27B) 66.6
  25. 25 Bespoke Nimble v4 Max (27B) 66.6
  26. 26 AnyJev Tacitus 31B 66.5
  27. 27 Decider 31B (NVFP4) 66.0
  28. 28 KnowLine 35B-A3B Gen1.5 65.8
  29. 29 OneShot 27B 65.7
  30. 30 vera-large 65.3
  31. 31 Bobcat 1.3 (31B) 64.8
  32. 32 Amos 1 Large (27B) 64.8
  33. 33 Matilda Decision v1 64.8
  34. 34 KnowLine 26B-A4B Gen2.5 64.6
  35. 35 frontier-infra Jebadiah 27B v2.1 (t1e) 64.4
  36. 36 deck31b (Gemma 4 31B) 63.0
  37. 37 Darwin-27B-ZTC 63.0
  38. 38 decisio 0.6.0 (Gemma 4 31B) 62.7
  39. 39 Cerno 27B 61.7
  40. 40 Torchcast Decision 27B 61.5
  41. 41 OneJev1.6-27B 61.5
  42. 42 Qwen Decision 27B noncausal LoRA 61.1
  43. 43 Quyet-1.0-Large 60.8
  44. 44 garum (Gemma 4 31B) 60.3
  45. 45 baselabs elephant-1 (9B) 60.3
  46. 46 Standard One 35B SH (v1) 60.1
  47. 47 Lumma-fev-27B 60.0
  48. 48 Sieve 27B v2 59.9
  49. 49 mogan-decision-31B-it 59.8
  50. 50 linne-fern 1.0 (27B) 59.6
  51. 51 Kev 27B 59.6
  52. 52 Xor 26B-A4B v1.2 59.4
  53. 53 Jev 1.13 59.2
  54. 54 span-02 (Gemma 4 26B-A4B) 59.1
  55. 55 garas (Gemma 4 31B) 59.0
  56. 56 Shisa DE-2.1 (repeat) 58.8
  57. 57 Eikos-27B 58.8
  58. 58 reflex Qwen3.8-27B-FP8 (pmscr) 58.8
  59. 59 RSI-Jev v6.1 VL 27B (effort auto) 58.6
  60. 60 KnowLine 9B Gen1 58.4
  61. 61 StartLux-Decision v1.1 9B 58.1
  62. 62 vLLM-SR Decision 2.0 Vega 27B 58.0
  63. 63 JEV-35B 57.6
  64. 64 JEV-27B 57.4
  65. 65 Qwen 3.8 27B 57.0
  66. 66 Tern-35B-A3B 56.6
  67. 67 Surogate Rune 26B-A4B v3 56.4
  68. 68 SynACK Decide 26B-A4B v7 56.2
  69. 69 KnowLine 4B Gen5.5 56.0
  70. 70 Blink v0.3 26B-A4B (NVFP4) 55.9
  71. 71 JADE 55.6
  72. 72 Wald-26B-A4B 55.4
  73. 73 AJev lora1 (Gemma 4 26B-A4B) 55.3
  74. 74 Cloudflare clef 54.9
  75. 75 StartLux-Decision v1.1 4B 54.8
  76. 76 Qwen3.6 27B 54.7
  77. 77 Perplexity Decider Flash v1.2 (4B) 54.3
  78. 78 vLLM-SR d3-flash (9B) 54.3
  79. 79 JPT-35B-A3B 54.1
  80. 80 Xor 1.2 53.6
  81. 81 GEV-26B-Decide 53.5
  82. 82 OneJev1.6-9B 53.0
  83. 83 Nace.AI Drex v2.0 (35B-A3B) 52.8
  84. 84 muria (Gemma 4 12B) 52.5
  85. 85 Jeffan-4B-2010 52.2
  86. 86 Verdin-4B v1.1 52.0
  87. 87 Bespoke Nimble 9B v3 51.1
  88. 88 Senova-4B v2 51.0
  89. 89 frontier-infra Jebadiah 9B v2.1 (t1e) 50.9
  90. 90 decisio 0.4.0 (Qwen3.6-35B-A3B) 50.8
  91. 91 vLLM-SR d3-mini (4B) 50.0
  92. 92 JoshuaSP diffusiongemma (open-jev) 49.5
  93. 93 AJev lora7 (Gemma 4 12B) 48.9
  94. 94 OneJev1.6-4B 48.9
  95. 95 decisio 0.4.0 (Gemma 4 12B) 48.8
  96. 96 Mica v0.5 4B 47.7
  97. 97 PurpleMIST-Flash-1.0 47.5
  98. 98 Cerno 9B v1.1 47.3
  99. 99 Sieve 9B v2 47.0
  100. 100 Decider 35B-A3B 46.6
  101. 101 Winnow-12B 46.6
  102. 102 Seb-9B v2 46.3
  103. 103 vLLM-SR Decision 2.0 Lux 9B 46.0
  104. 104 Nace.AI Drex v1.5 45.8
  105. 105 Hopper 12B (trained) 45.8
  106. 106 Trio-Spark DI (9B) 45.6
  107. 107 CoCo-Decision-4B 2.4 45.5
  108. 108 Cloudflare clef-omni (30B-A3B) 45.0
  109. 109 JevK5 9B v0.3.3 44.9
  110. 110 Cloudflare clef-flash 44.4
  111. 111 Jevfire 44.4
  112. 112 RSI-Jev v6.1 VL 4B 44.3
  113. 113 Jegue 0.1 (9B) 44.2
  114. 114 jiwo 4B v2 44.1
  115. 115 JPT-9B 44.1
  116. 116 ezjev 4B s2 43.9
  117. 117 Jev-Omni 43.8
  118. 118 Solomon v1.1 43.2
  119. 119 Standard One 8B SH (v4) 42.9
  120. 120 Amber Hybrid 4B 42.7
  121. 121 Saif 4B 42.5
  122. 122 JEV-9B 42.1
  123. 123 Bespoke Nimble 9B v2 42.1
  124. 124 Sieve 9B 41.7
  125. 125 Cerno 4B v1.1 41.1
  126. 126 Sieve 4B 41.0
  127. 127 Kev 9B v2 40.9
  128. 128 Nari Decision 4B v3.2 40.9
  129. 129 Aplomb 1 40.8
  130. 130 vLLM-SR Decision 2.0 Nox 4B 40.8
  131. 131 Hopper (G) 1.2 40.2
  132. 132 StartLux-Decision v1.1 2B 40.1
  133. 133 FlyMy.AI Decision 4B v1.6 39.8
  134. 134 Standard One 3B SH (v4) 39.8
  135. 135 JPT-4B 39.6
  136. 136 Amber 4B 39.5
  137. 137 Intern-Decision-4B 38.7
  138. 138 RSI-Jev v5.0 VL 3B 38.6
  139. 139 djev 38.1
  140. 140 Valen-4B 37.9
  141. 141 razorback16 openjev diffusiongemma (NVFP4, vLLM) 37.6
  142. 142 NeoHorse-Jev-4B 37.3
  143. 143 JevK5 v0.3 36.6
  144. 144 Decider 4B 36.3
  145. 145 baselabs joey-1 (2B) 35.9
  146. 146 lev 35.5
  147. 147 Lev1 3.7B 35.4
  148. 148 Kev 4B r10 34.9
  149. 149 Jev-Style 2B v4 34.5
  150. 150 Sieve 2B v3 34.5
  151. 151 LatentMesh Blitz 2B v1 34.4
  152. 152 Saracura PT-BR 4B 34.4
  153. 153 Cam1 34.4
  154. 154 vLLM-SR d3-nano (2B) 34.3
  155. 155 mmastrac diffusiongemma (vLLM PR 57250) 34.3
  156. 156 Kas 4B 34.0
  157. 157 Tev1-4B-experimental 33.7
  158. 158 Winnow-E4B 32.7
  159. 159 Aurelis-2B 32.5
  160. 160 Jevstral 8B 32.2
  161. 161 Metask-Jev-4B 32.2
  162. 162 Ryotide Qwen v0.3.1 32.0
  163. 163 Intelif Qwen3-4B 31.3
  164. 164 GLiClass v4 edge (0.4B) 31.2
  165. 165 LiquidAI d1-3B 31.1
  166. 166 open-jev (pngwn) 31.1
  167. 167 vera-core-2 30.6
  168. 168 Jobe 30.4
  169. 169 OneJev1.6-0.8B 29.6
  170. 170 crystal-0.8b 29.4
  171. 171 vLLM-SR vela3-fast (Vela 3.0, 0.4B) 29.3
  172. 172 Crivo-1B 29.3
  173. 173 Gevva E4B 29.2
  174. 174 StartLux-Decision v1.1 0.8B 28.9
  175. 175 openvons 28.8
  176. 176 SemIf 26.7
  177. 177 Saina Helm 2 Omni (0.8B) 26.5
  178. 178 vLLM-SR d3-lite (0.8B) 26.2
  179. 179 Amber 0.4B 26.0
  180. 180 jiwo 0.8B v2 25.8
  181. 181 Saina Helm 2.0 v2 25.3
  182. 182 this-that 1.2 24.9
  183. 183 Gevva E2B 24.3
  184. 184 Crivo-400M 23.7
  185. 185 Decider 2B 23.6
  186. 186 vLLM-SR Decision 2.0 Sol 2B 23.6
  187. 187 jdi-ptr14-0.8b 23.6
  188. 188 mini-jev 22.8
  189. 189 Valen-2B 22.2
  190. 190 vLLM-SR vela3-fast-vision (Vela 3.0, 0.25B) 22.1
  191. 191 Sieve 2B 21.9
  192. 192 wyat-1-800m 21.5
  193. 193 Intern-Decision-2B 20.9
  194. 194 Sieve 0.8B 20.8
  195. 195 Strands Decider 2B 20.7
  196. 196 EXAONE-4.0-1.2B-JEV v0.3 20.6
  197. 197 tasksource Decider nano v5 (150M) 19.9
  198. 198 Crivo-68M 19.5
  199. 199 JEV Ettin 150M 18.5
  200. 200 vLLM-SR Decision 2.0 Eos 0.8B 18.4
  201. 201 Crivo-150M v2 18.3
  202. 202 Bekko System One v0 400M 18.0
  203. 203 Mola-150M 17.8
  204. 204 vLLM-SR d3-edge (0.6B) 17.7
  205. 205 Mola-68M 17.6
  206. 206 JEV FLAN-T5 Large public v2 (342M) 17.3
  207. 207 DecisionTune 1.0 17.2
  208. 208 Valen-0.8B 16.6
  209. 209 Bosun v3.1 1.7B 16.5
  210. 210 vLLM-SR Decision 2.0 Kai 0.6B 16.4
  211. 211 Dinah-1 16.4
  212. 212 Kev 0.8B r15 16.3
  213. 213 PICK-150M v5 15.6
  214. 214 JPT-0.8B 15.5
  215. 215 JEV Ettin 400M 15.2
  216. 216 Sifr 0.8B v3.1 14.5
  217. 217 Crivo-32M v2 14.2
  218. 218 Prumo 14.0
  219. 219 Kodiak v0.2 1B (accuracy) 13.9
  220. 220 Intern-Decision-0.8B 12.7
  221. 221 Jev-Style 0.8B v3 12.5
  222. 222 Tev1-0.8B-experimental 12.5
  223. 223 Bekko System One v0 68M 12.4
  224. 224 LiquidAI d1-omni-600M 11.8
  225. 225 Crivo-17M 11.7
  226. 226 Von 1.3 11.7
  227. 227 Bosun v3.1 0.6B 11.0
  228. 228 vera-spark 11.0
  229. 229 Lavoir 10.8
  230. 230 GLiNER2.5-Decide 10.1
  231. 231 Laya 9.2
  232. 232 JevK5 Lite 8.8
  233. 233 LFM2.5-2.6B-RLCD 8.2
  234. 234 Qwen-2.5-1B-RLCD 7.8
  235. 235 Bekko System One v0 17M 7.7
  236. 236 CLM-v0.1-8B 6.5
  237. 237 jeff 6.2
  238. 238 LFM2.5-350M-RLCD 5.9
  239. 239 system-one-gemma 5.2
  240. 240 DM-JEPA 1.1 5.0
  241. 241 Julia 1 4.8
  242. 242 Nirnaya 0.4B 3.5
  243. 243 Zev 3.4
  244. 244 Verdict 2.0
#EntryScoreSettingDateSource
1 KnowLine 31B Gen1 71.9 Official leaderboard
2 Fastino GLiDE 1.2 71.5 Official leaderboard
3 Sieve 31B v3 71.1 Official leaderboard
4 Turion Jev 31B 70.5 Official leaderboard
5 vLLM-SR d3-ultra (Decision 3.0, 27B) 70.2 Official leaderboard
6 Jeffan-31b-1710 69.8 Official leaderboard
7 H2O Lightning 31B v1.1 68.6 Official leaderboard
8 WeMM-JEV-27B 68.5 Official leaderboard
9 baselabs mammoth-1 (27B) 68.4 Official leaderboard
10 Ainvest-Decision-27B v1.0 68.2 Official leaderboard
11 oenogarum (Gemma 4 31B) 68.1 Official leaderboard
12 Perplexity Decider v1.2 (27B) 68.0 Official leaderboard
13 AutoTrust-JEV-27B v1.1 68.0 Official leaderboard
14 StartLux-Decision v1.1 27B 67.9 Official leaderboard
15 MISA AI 2.0 Decision Plus · oak 1.1 (31B) 67.6 Official leaderboard
16 AJev lm6q (Gemma 4 31B) 67.6 Official leaderboard
17 Gemma 4 31B IT 67.6 FlexDM Apr 2026 Official leaderboard
18 Lychee-27B 67.3 Official leaderboard
19 vLLM-SR d3-turbo v1.1 (27B) 67.3 Official leaderboard
20 vLLM-SR d3 (27B) 67.2 Official leaderboard
21 allec-oavg (Gemma 4 31B) 67.0 Official leaderboard
22 VisionJEV 1.0.0 (27B) 66.9 Official leaderboard
23 Aleria Judge 1.1.1 (Gemma 4 31B) 66.7 Official leaderboard
24 Nace.AI Drex Max v1 (27B) 66.6 Official leaderboard
25 Bespoke Nimble v4 Max (27B) 66.6 Official leaderboard
26 AnyJev Tacitus 31B 66.5 Official leaderboard
27 Decider 31B (NVFP4) 66.0 Official leaderboard
28 KnowLine 35B-A3B Gen1.5 65.8 Official leaderboard
29 OneShot 27B 65.7 Official leaderboard
30 vera-large 65.3 Official leaderboard
31 Bobcat 1.3 (31B) 64.8 Official leaderboard
32 Amos 1 Large (27B) 64.8 Official leaderboard
33 Matilda Decision v1 64.8 Official leaderboard
34 KnowLine 26B-A4B Gen2.5 64.6 Official leaderboard
35 frontier-infra Jebadiah 27B v2.1 (t1e) 64.4 Official leaderboard
36 deck31b (Gemma 4 31B) 63.0 Official leaderboard
37 Darwin-27B-ZTC 63.0 Official leaderboard
38 decisio 0.6.0 (Gemma 4 31B) 62.7 Official leaderboard
39 Cerno 27B 61.7 Official leaderboard
40 Torchcast Decision 27B 61.5 Official leaderboard
41 OneJev1.6-27B 61.5 Official leaderboard
42 Qwen Decision 27B noncausal LoRA 61.1 Official leaderboard
43 Quyet-1.0-Large 60.8 Official leaderboard
44 garum (Gemma 4 31B) 60.3 Official leaderboard
45 baselabs elephant-1 (9B) 60.3 Official leaderboard
46 Standard One 35B SH (v1) 60.1 Official leaderboard
47 Lumma-fev-27B 60.0 Official leaderboard
48 Sieve 27B v2 59.9 Official leaderboard
49 mogan-decision-31B-it 59.8 Official leaderboard
50 linne-fern 1.0 (27B) 59.6 Official leaderboard
51 Kev 27B 59.6 Official leaderboard
52 Xor 26B-A4B v1.2 59.4 Official leaderboard
53 Jev 1.13 59.2 Official leaderboard
54 span-02 (Gemma 4 26B-A4B) 59.1 Official leaderboard
55 garas (Gemma 4 31B) 59.0 Official leaderboard
56 Shisa DE-2.1 (repeat) 58.8 Official leaderboard
57 Eikos-27B 58.8 Official leaderboard
58 reflex Qwen3.8-27B-FP8 (pmscr) 58.8 Official leaderboard
59 RSI-Jev v6.1 VL 27B (effort auto) 58.6 Official leaderboard
60 KnowLine 9B Gen1 58.4 Official leaderboard
61 StartLux-Decision v1.1 9B 58.1 Official leaderboard
62 vLLM-SR Decision 2.0 Vega 27B 58.0 Official leaderboard
63 JEV-35B 57.6 Official leaderboard
64 JEV-27B 57.4 Official leaderboard
65 Qwen 3.8 27B 57.0 simple-jev Aug 2026 Official leaderboard
66 Tern-35B-A3B 56.6 Official leaderboard
67 Surogate Rune 26B-A4B v3 56.4 Official leaderboard
68 SynACK Decide 26B-A4B v7 56.2 Official leaderboard
69 KnowLine 4B Gen5.5 56.0 Official leaderboard
70 Blink v0.3 26B-A4B (NVFP4) 55.9 Official leaderboard
71 JADE 55.6 Official leaderboard
72 Wald-26B-A4B 55.4 Official leaderboard
73 AJev lora1 (Gemma 4 26B-A4B) 55.3 Official leaderboard
74 Cloudflare clef 54.9 Official leaderboard
75 StartLux-Decision v1.1 4B 54.8 Official leaderboard
76 Qwen3.6 27B 54.7 Decider chat Apr 2026 Official leaderboard
77 Perplexity Decider Flash v1.2 (4B) 54.3 Official leaderboard
78 vLLM-SR d3-flash (9B) 54.3 Official leaderboard
79 JPT-35B-A3B 54.1 Official leaderboard
80 Xor 1.2 53.6 Official leaderboard
81 GEV-26B-Decide 53.5 Official leaderboard
82 OneJev1.6-9B 53.0 Official leaderboard
83 Nace.AI Drex v2.0 (35B-A3B) 52.8 Official leaderboard
84 muria (Gemma 4 12B) 52.5 Official leaderboard
85 Jeffan-4B-2010 52.2 Official leaderboard
86 Verdin-4B v1.1 52.0 Official leaderboard
87 Bespoke Nimble 9B v3 51.1 Official leaderboard
88 Senova-4B v2 51.0 Official leaderboard
89 frontier-infra Jebadiah 9B v2.1 (t1e) 50.9 Official leaderboard
90 decisio 0.4.0 (Qwen3.6-35B-A3B) 50.8 Official leaderboard
91 vLLM-SR d3-mini (4B) 50.0 Official leaderboard
92 JoshuaSP diffusiongemma (open-jev) 49.5 Official leaderboard
93 AJev lora7 (Gemma 4 12B) 48.9 Official leaderboard
94 OneJev1.6-4B 48.9 Official leaderboard
95 decisio 0.4.0 (Gemma 4 12B) 48.8 Official leaderboard
96 Mica v0.5 4B 47.7 Official leaderboard
97 PurpleMIST-Flash-1.0 47.5 Official leaderboard
98 Cerno 9B v1.1 47.3 Official leaderboard
99 Sieve 9B v2 47.0 Official leaderboard
100 Decider 35B-A3B 46.6 Official leaderboard
101 Winnow-12B 46.6 Official leaderboard
102 Seb-9B v2 46.3 Official leaderboard
103 vLLM-SR Decision 2.0 Lux 9B 46.0 Official leaderboard
104 Nace.AI Drex v1.5 45.8 Official leaderboard
105 Hopper 12B (trained) 45.8 Official leaderboard
106 Trio-Spark DI (9B) 45.6 Official leaderboard
107 CoCo-Decision-4B 2.4 45.5 Official leaderboard
108 Cloudflare clef-omni (30B-A3B) 45.0 Official leaderboard
109 JevK5 9B v0.3.3 44.9 Official leaderboard
110 Cloudflare clef-flash 44.4 Official leaderboard
111 Jevfire 44.4 Official leaderboard
112 RSI-Jev v6.1 VL 4B 44.3 Official leaderboard
113 Jegue 0.1 (9B) 44.2 Official leaderboard
114 jiwo 4B v2 44.1 Official leaderboard
115 JPT-9B 44.1 Official leaderboard
116 ezjev 4B s2 43.9 Official leaderboard
117 Jev-Omni 43.8 Official leaderboard
118 Solomon v1.1 43.2 Official leaderboard
119 Standard One 8B SH (v4) 42.9 Official leaderboard
120 Amber Hybrid 4B 42.7 Official leaderboard
121 Saif 4B 42.5 Official leaderboard
122 JEV-9B 42.1 Official leaderboard
123 Bespoke Nimble 9B v2 42.1 Official leaderboard
124 Sieve 9B 41.7 Official leaderboard
125 Cerno 4B v1.1 41.1 Official leaderboard
126 Sieve 4B 41.0 Official leaderboard
127 Kev 9B v2 40.9 Official leaderboard
128 Nari Decision 4B v3.2 40.9 Official leaderboard
129 Aplomb 1 40.8 Official leaderboard
130 vLLM-SR Decision 2.0 Nox 4B 40.8 Official leaderboard
131 Hopper (G) 1.2 40.2 Official leaderboard
132 StartLux-Decision v1.1 2B 40.1 Official leaderboard
133 FlyMy.AI Decision 4B v1.6 39.8 Official leaderboard
134 Standard One 3B SH (v4) 39.8 Official leaderboard
135 JPT-4B 39.6 Official leaderboard
136 Amber 4B 39.5 Official leaderboard
137 Intern-Decision-4B 38.7 Official leaderboard
138 RSI-Jev v5.0 VL 3B 38.6 Official leaderboard
139 djev 38.1 Official leaderboard
140 Valen-4B 37.9 Official leaderboard
141 razorback16 openjev diffusiongemma (NVFP4, vLLM) 37.6 Official leaderboard
142 NeoHorse-Jev-4B 37.3 Official leaderboard
143 JevK5 v0.3 36.6 Official leaderboard
144 Decider 4B 36.3 Official leaderboard
145 baselabs joey-1 (2B) 35.9 Official leaderboard
146 lev 35.5 Official leaderboard
147 Lev1 3.7B 35.4 Official leaderboard
148 Kev 4B r10 34.9 Official leaderboard
149 Jev-Style 2B v4 34.5 Official leaderboard
150 Sieve 2B v3 34.5 Official leaderboard
151 LatentMesh Blitz 2B v1 34.4 Official leaderboard
152 Saracura PT-BR 4B 34.4 Official leaderboard
153 Cam1 34.4 Official leaderboard
154 vLLM-SR d3-nano (2B) 34.3 Official leaderboard
155 mmastrac diffusiongemma (vLLM PR 57250) 34.3 Official leaderboard
156 Kas 4B 34.0 Official leaderboard
157 Tev1-4B-experimental 33.7 Official leaderboard
158 Winnow-E4B 32.7 Official leaderboard
159 Aurelis-2B 32.5 Official leaderboard
160 Jevstral 8B 32.2 Official leaderboard
161 Metask-Jev-4B 32.2 Official leaderboard
162 Ryotide Qwen v0.3.1 32.0 Official leaderboard
163 Intelif Qwen3-4B 31.3 Official leaderboard
164 GLiClass v4 edge (0.4B) 31.2 Official leaderboard
165 LiquidAI d1-3B 31.1 Official leaderboard
166 open-jev (pngwn) 31.1 Official leaderboard
167 vera-core-2 30.6 Official leaderboard
168 Jobe 30.4 Official leaderboard
169 OneJev1.6-0.8B 29.6 Official leaderboard
170 crystal-0.8b 29.4 Official leaderboard
171 vLLM-SR vela3-fast (Vela 3.0, 0.4B) 29.3 Official leaderboard
172 Crivo-1B 29.3 Official leaderboard
173 Gevva E4B 29.2 Official leaderboard
174 StartLux-Decision v1.1 0.8B 28.9 Official leaderboard
175 openvons 28.8 Official leaderboard
176 SemIf 26.7 Official leaderboard
177 Saina Helm 2 Omni (0.8B) 26.5 Official leaderboard
178 vLLM-SR d3-lite (0.8B) 26.2 Official leaderboard
179 Amber 0.4B 26.0 Official leaderboard
180 jiwo 0.8B v2 25.8 Official leaderboard
181 Saina Helm 2.0 v2 25.3 Official leaderboard
182 this-that 1.2 24.9 Official leaderboard
183 Gevva E2B 24.3 Official leaderboard
184 Crivo-400M 23.7 Official leaderboard
185 Decider 2B 23.6 Official leaderboard
186 vLLM-SR Decision 2.0 Sol 2B 23.6 Official leaderboard
187 jdi-ptr14-0.8b 23.6 Official leaderboard
188 mini-jev 22.8 Official leaderboard
189 Valen-2B 22.2 Official leaderboard
190 vLLM-SR vela3-fast-vision (Vela 3.0, 0.25B) 22.1 Official leaderboard
191 Sieve 2B 21.9 Official leaderboard
192 wyat-1-800m 21.5 Official leaderboard
193 Intern-Decision-2B 20.9 Official leaderboard
194 Sieve 0.8B 20.8 Official leaderboard
195 Strands Decider 2B 20.7 Official leaderboard
196 EXAONE-4.0-1.2B-JEV v0.3 20.6 Official leaderboard
197 tasksource Decider nano v5 (150M) 19.9 Official leaderboard
198 Crivo-68M 19.5 Official leaderboard
199 JEV Ettin 150M 18.5 Official leaderboard
200 vLLM-SR Decision 2.0 Eos 0.8B 18.4 Official leaderboard
201 Crivo-150M v2 18.3 Official leaderboard
202 Bekko System One v0 400M 18.0 Official leaderboard
203 Mola-150M 17.8 Official leaderboard
204 vLLM-SR d3-edge (0.6B) 17.7 Official leaderboard
205 Mola-68M 17.6 Official leaderboard
206 JEV FLAN-T5 Large public v2 (342M) 17.3 Official leaderboard
207 DecisionTune 1.0 17.2 Official leaderboard
208 Valen-0.8B 16.6 Official leaderboard
209 Bosun v3.1 1.7B 16.5 Official leaderboard
210 vLLM-SR Decision 2.0 Kai 0.6B 16.4 Official leaderboard
211 Dinah-1 16.4 Official leaderboard
212 Kev 0.8B r15 16.3 Official leaderboard
213 PICK-150M v5 15.6 Official leaderboard
214 JPT-0.8B 15.5 Official leaderboard
215 JEV Ettin 400M 15.2 Official leaderboard
216 Sifr 0.8B v3.1 14.5 Official leaderboard
217 Crivo-32M v2 14.2 Official leaderboard
218 Prumo 14.0 Official leaderboard
219 Kodiak v0.2 1B (accuracy) 13.9 Official leaderboard
220 Intern-Decision-0.8B 12.7 Official leaderboard
221 Jev-Style 0.8B v3 12.5 Official leaderboard
222 Tev1-0.8B-experimental 12.5 Official leaderboard
223 Bekko System One v0 68M 12.4 Official leaderboard
224 LiquidAI d1-omni-600M 11.8 Official leaderboard
225 Crivo-17M 11.7 Official leaderboard
226 Von 1.3 11.7 Official leaderboard
227 Bosun v3.1 0.6B 11.0 Official leaderboard
228 vera-spark 11.0 Official leaderboard
229 Lavoir 10.8 Official leaderboard
230 GLiNER2.5-Decide 10.1 Official leaderboard
231 Laya 9.2 Official leaderboard
232 JevK5 Lite 8.8 Official leaderboard
233 LFM2.5-2.6B-RLCD 8.2 Official leaderboard
234 Qwen-2.5-1B-RLCD 7.8 Official leaderboard
235 Bekko System One v0 17M 7.7 Official leaderboard
236 CLM-v0.1-8B 6.5 Official leaderboard
237 jeff 6.2 Official leaderboard
238 LFM2.5-350M-RLCD 5.9 Official leaderboard
239 system-one-gemma 5.2 Official leaderboard
240 DM-JEPA 1.1 5.0 Official leaderboard
241 Julia 1 4.8 Official leaderboard
242 Nirnaya 0.4B 3.5 Official leaderboard
243 Zev 3.4 Official leaderboard
244 Verdict 2.0 Official leaderboard

Number one over time

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

Google DeepMind
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
50 55 60 65 70 2026 50 55 60 65 70 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 Other

All 41 in Other
  • Personal Superintelligence Index (Meta) In use Meta's composite of seven personal and knowledge-work evals: Agents' Last Exam, AutomationBench-AA, FocusBench, JobBench, GDP.pdf, VBT and DRACO.
  • Artificial Analysis Intelligence Index v5 In use Version 5 of Artificial Analysis's index: nine evaluations in four equal categories, most with private test sets, run three times each in its own agent harness. Gemini 4 Carbon 59.0
  • Anthropic ECI (AECI) Only in system cards Index that combines a basket of capability benchmarks into one capability score to track trends across Anthropic models. Claude Opus 5.5 169.4
  • Qwen in-house English automatic evaluation Only in system cards Models answer Qwen in-house English datasets on knowledge, comprehension, coding, math and related skills. Skills and weights change between reports. Qwen2-72B 72.9%
  • Recognition of model-written user turns (Anthropic) Only in system cards Models write synthetic user turns for real Claude Code transcripts, then judge which version is real. Score is accuracy above chance. Claude Mythos Preview 0.35
  • LiveBench Dropped Models answer monthly refreshed questions across math, coding, reasoning, language, instruction following, and data analysis. Scored automatically against objective ground truth. Claude Fable 5.1 85.7%