Benchmarks
More
OverviewBenchmaxxingCompareChartsCost vs CapabilityProviders per ModelProvider explorerGatewaysEU & SovereignJev-class modelsImage JevBenchAudioJevBenchSubmit a modelAbout
JevBench v1.8.11 · method v1.8 (10 s rule, GPU-time cost basis)

JevBench v1.8.11: Minimodels Decision 1.0 (candy-cane)

Added Minimodels Decision 1.0 (candy-cane), measured on the 27 Oct draw, unranked until the next ranking review.

Aggregate results JSON · SHA-256 3599bbf28797e1da9a50de887ff500a2ffa8643cd2001e9ab3e635e08b592b69 · Previous v1.8.10

JevBench v1.8.11 · headline

JevBench Capability Score

Capability ranking of Jev-class systems

Capability Score averages Intelligence and Calibration. Torchcast Decision 31B v1.0 leads the Jev-class systems with 89.5.

Jev-class means at most 2× the cost and median latency of Jev (v1.5 reference). How we choose ↘

Adjust cost / latency caps · 2× official
Official
#SystemICCap.$/1k
  1. 1Torchcast Decision 31B v1.0Usev1.7.14 drawv1.7 · 10 s rule85.559.489.5$0.023*Use this model
  2. 2MetaDecision-31B-RainUsev1.8.6 drawv1.7 · 10 s rule85.055.088.3$0.032*Use this model
  3. 3Karpo-Decision-31BUsev1.8.9 drawv1.7 · 10 s rule80.257.987.2$0.025*Use this model
  4. 4kettei-fatAPIv1.6.43 drawv1.7 · 10 s rule79.857.486.3$0.026*
  5. 5Super DeciderAPIv1.7.0 drawv1.7 · 10 s rule80.056.485.9$0.028
  6. 6baselabs mammoth-1Usev1.7.0 drawv1.7 · 10 s rule76.257.285.2$0.027*Use this model
  7. 7KnowLine 31B-Gen1Usev1.7.0 drawv1.7 · 10 s rule77.446.585.2$0.061*Use this model
  8. 8Fastino GLiDE 1.2Usev1.7.0 drawv1.7 · 10 s rule76.458.184.9$0.025*Use this model
  9. 9Super DeciderAPIv1.7.0 drawv1.7 · 10 s rule78.964.384.9$0.016
  10. 10Fastino GLiDEAPIv1.6.43 drawv1.7 · 10 s rule76.159.184.5$0.023
Show all 282 Jev-class systems (272 more)
  1. 11H2O-Lightning-31B v1.0.7Usev1.6.43 drawv1.7 · 10 s rule79.356.084.3$0.029*Use this model
  2. 12Drex MaxAPIv1.6.43 drawv1.7 · 10 s rule78.955.384.2$0.031
  3. 13Piccolo 27BUsev1.8.3 drawv1.7 · 10 s rule75.853.183.6$0.037*Use this model
  4. 14Perplexity Decider v1.1 27BUsev1.7 · 10 s rule75.657.382.7$0.026*Use this model
  5. 15vLLM-SR d3-turboUsev1.6.43 drawv1.7 · 10 s rule74.857.382.6$0.027*Use this model
  6. 16StartLux-Decision v1.1 27B InstantUsev1.6.43 drawv1.7 · 10 s rule73.257.982.6$0.025*Use this model
  7. 17vLLM-SR d3Usev1.6.43 drawv1.7 · 10 s rule72.957.382.1$0.027*Use this model
  8. 18basal-omni-1.0-ultraUsev1.6.43 drawv1.7 · 10 s rule74.155.982.1$0.029*Use this model
  9. 19DOPEST 1.0 31BUsev1.7.4 drawv1.7 · 10 s rule73.952.682.0$0.038*Use this model
  10. 20SubQ Snap 1.1APIv1.6.43 drawv1.7 · 10 s rule75.860.781.9$0.020*
  11. 21Jeff-1.1-MediumUsev1.7.8 drawv1.7 · 10 s rule74.566.281.9$0.013*Use this model
  12. 22Arcbright Elf 1 31BUsev1.7.0 drawv1.7 · 10 s rule71.955.181.8$0.031*Use this model
  13. 23Quyet-1.0-LargeUsev1.7 · 10 s rule73.455.981.7$0.030*Use this model
  14. 24Jeffan-31b-1710Usev1.7.0 drawv1.7 · 10 s rule71.651.481.7$0.042*Use this model
  15. 25decide-1.1APIv1.8.1 drawv1.7 · 10 s rule72.870.281.6$0.0098
  16. 26oenogarumUsev1.8.2 drawv1.7 · 10 s rule71.855.981.5$0.030*Use this model
  17. 27Jeff-1.0-LargeUsev1.6.4 drawv1.7 · 10 s rule71.954.581.5$0.033*Use this model
  18. 28SubQ SnapAPIv1.6.43 drawv1.7 · 10 s rule72.357.781.5$0.026*
  19. 29Nimble-v4-maxUsev1.7.0 drawv1.7 · 10 s rule71.057.581.2$0.026*Use this model
  20. 30Sieve-31B-v2Usev1.7.5 drawv1.7 · 10 s rule68.756.281.1$0.029*Use this model
  21. 31Turion Jev 31BUsev1.7.1 drawv1.7 · 10 s rule70.946.680.9$0.060*Use this model
  22. 32pplx-decider-v1.2Usev1.6.43 drawv1.7 · 10 s rule71.757.480.8$0.026*Use this model
  23. 33allec-oavgUsev1.8.3 drawv1.7 · 10 s rule70.555.980.8$0.030*Use this model
  24. 34d1-ultraAPIv1.6.43 drawv1.7 · 10 s rule69.762.380.5$0.018*
  25. 35torchcast-decision-27bUsev1.7 · 10 s rule70.457.780.2$0.026*Use this model
  26. 36Decision-King v1.1Usev1.8.1 drawv1.7 · 10 s rule72.861.179.9$0.020*Use this model
  27. 37Lychee-27BUsev1.7.20 drawv1.7 · 10 s rule74.357.379.8$0.026*Use this model
  28. 38Bespoke Nimble-v4-maxAPIv1.6.43 drawv1.7 · 10 s rule68.260.779.7$0.020*
  29. 39metask-jev-rain-31B-trained service v3Usev1.6.43 drawv1.7 · 10 s rule72.655.479.6$0.031*Use this model
  30. 40decisio v0.8.0 on gemma-4-31B-itUsev1.7 · 10 s rule70.455.079.6$0.032*Use this model
  31. 41Vera LargeUsev1.7.0 drawv1.7 · 10 s rule67.855.979.6$0.029*Use this model
  32. 42Kev 27BUsev1.7 · 10 s rule64.256.278.9$0.029*Use this model
  33. 43metask-jev-rain-31B-trainedUsev1.6.43 drawv1.7 · 10 s rule70.756.178.9$0.029*Use this model
  34. 44Myles-G 31BAPIv1.6.43 drawv1.7 · 10 s rule73.651.178.8$0.043*
  35. 45StartLux-Decision v1.1 27B AutoThinkUsev1.6.43 drawv1.7 · 10 s rule66.756.678.8$0.028*Use this model
  36. 46Sage 1.3.0APIv1.7 · 10 s rule65.658.278.6$0.025*
  37. 47Gemini Decision 1APIv1.8.10 drawv1.7 · 10 s rule67.452.578.5$0.038
  38. 48Bobcat 1.3Usev1.7.0 drawv1.7 · 10 s rule68.155.277.9$0.031*Use this model
  39. 49deck-31BUsev1.7 · 10 s rule73.055.077.6$0.032*Use this model
  40. 50Torchcast Decision 4BUsev1.6.43 drawv1.7 · 10 s rule67.179.477.5$0.0049*Use this model
  41. 51Arcbright Elf 1 12BUsev1.7.14 drawv1.7 · 10 s rule63.971.777.5$0.0088*Use this model
  42. 52Microsoft-Decision-1APIv1.7.15 drawv1.7 · 10 s rule64.266.477.4$0.013
  43. 53KnowLine 26B-A4B-Gen2.5Usev1.7.18 drawv1.7 · 10 s rule65.262.377.3$0.018*Use this model
  44. 54Jeff-1.1-Qwen-4BUsev1.7.14 drawv1.7 · 10 s rule65.380.777.3$0.0044*Use this model
  45. 55Eikos-27BUsev1.7 · 10 s rule64.456.177.2$0.029*Use this model
  46. 56Jev 1.13.0APIv1.7 · 10 s rule63.654.777.1$0.032
  47. 57metask-jev-rain-12B-trainedUsev1.6.43 drawv1.7 · 10 s rule66.565.276.9$0.014*Use this model
  48. 58Wald 26B-A4BUsev1.7.0 drawv1.7 · 10 s rule65.571.076.8$0.0093*Use this model
  49. 59KnowLine 35B-A3B-Gen1.5Usev1.7.0 drawv1.7 · 10 s rule60.363.676.7$0.016*Use this model
  50. 60Decisio 12B tuned commit-thinkUsev1.7.14 drawv1.7 · 10 s rule65.770.776.3$0.0095*Use this model
  51. 61René-1 31B FP8Usev1.7 · 10 s rule61.751.376.0$0.042*Use this model
  52. 62Decisio 12B tuned commitUsev1.7.6 drawv1.7 · 10 s rule62.867.075.7$0.013*Use this model
  53. 63ClefUsev1.7 · 10 s rule59.155.575.3$0.030*Use this model
  54. 64SD-JEV-4B-v12.1.0APIv1.8.8 drawv1.7 · 10 s rule63.383.675.2$0.0035*
  55. 65Jev 1.14.0APIv1.6.43 drawv1.7 · 10 s rule59.954.875.0$0.032*
  56. 66H2O-Lightning-4B v1.1Usev1.6.43 drawv1.7 · 10 s rule60.078.675.0$0.0052*Use this model
  57. 67d1APIv1.7 · 10 s rule62.162.874.4$0.017
  58. 68SPX-CD FlashAPIv1.7 · 10 s rule58.452.174.3$0.039*
  59. 69Piccolo 9BUsev1.8.2 drawv1.7 · 10 s rule60.963.974.3$0.016*Use this model
  60. 70cribl-decision-1.0-27BUsev1.8.3 drawv1.7 · 10 s rule58.664.074.1$0.016*Use this model
  61. 71AutoTrust-Decider-Pro-27BUsev1.7 · 10 s rule56.858.673.9$0.024*Use this model
  62. 72Nagi-ENORMOUS 27BUsev1.6.43 drawv1.7 · 10 s rule62.457.473.8$0.026*Use this model
  63. 73Mercury DecideAPIv1.7 · 10 s rule65.962.173.6$0.018*
  64. 74Celeris-1 DecisionAPIv1.6.43 drawv1.7 · 10 s rule58.960.473.6$0.021*
  65. 75OpenAI DecisionsAPIv1.7 · 10 s rule56.948.673.5$0.052
  66. 76Surogate Rune 26B-A4B v3Usev1.7 · 10 s rule56.171.073.5$0.0092*Use this model
  67. 77RSI-Jev v6.1-VL 27BUsev1.6.9 drawv1.7 · 10 s rule61.658.773.5$0.024*Use this model
  68. 78kettei-chubAPIv1.6.43 drawv1.7 · 10 s rule62.262.773.1$0.018*
  69. 79Bobcat Flash 1.2Usev1.7 · 10 s rule65.671.173.0$0.0092*Use this model
  70. 80Torchcast Decision 4BUsev1.6.43 drawv1.7 · 10 s rule60.379.472.6$0.0049*Use this model
  71. 81swanOneUsev1.7 · 10 s rule56.271.472.4$0.0090*Use this model
  72. 82FrontiersMind Lumma-fev-27BUsev1.7.0 drawv1.7 · 10 s rule55.554.872.3$0.032*Use this model
  73. 83Umpire 35B-A3BUsev1.7.22 drawv1.7 · 10 s rule55.375.472.1$0.0066*Use this model
  74. 84z3-mini-soup-previewUsev1.8.3 drawv1.7 · 10 s rule57.582.872.1$0.0037*Use this model
  75. 85torchcast-decision-12bUsev1.7 · 10 s rule60.556.471.7$0.028*Use this model
  76. 86AutoJev-27BUsev1.7 · 10 s rule55.656.771.6$0.028*Use this model
  77. 87iota-decide-4b-v0.4Usev1.8.1 drawv1.7 · 10 s rule57.383.071.6$0.0037*Use this model
  78. 88decider-12b v2Usev1.7 · 10 s rule61.968.271.3$0.012*Use this model
  79. 89Decision 2.0 Vega 27BUsev1.7 · 10 s rule60.457.071.3$0.027*Use this model
  80. 90Jev-OmniUsev1.7 · 10 s rule55.566.571.3$0.013*Use this model
  81. 91Winnow-12B Q8Usev1.7 · 10 s rule59.567.071.2$0.013*Use this model
  82. 92Piccolo 4BUsev1.8.2 drawv1.7 · 10 s rule56.374.571.2$0.0071*Use this model
  83. 93Xor 26B-A4BUsev1.7 · 10 s rule57.572.171.2$0.0085*Use this model
  84. 94decider-12b v1, stock Gemma-4-12B-itUsev1.7 · 10 s rule59.668.271.0$0.012*Use this model
  85. 95CygnetUsev1.7 · 10 s rule54.866.870.9$0.013*Use this model
  86. 96Microsoft-Decision-1APIv1.7 · 10 s rule57.261.570.8$0.019
  87. 97decisio v0.8.0 on gemma-4-12B-itUsev1.7 · 10 s rule52.367.770.7$0.012*Use this model
  88. 98JPT-35B-A3BUsev1.7 · 10 s rule51.472.570.7$0.0083*Use this model
  89. 99Myles-G 12BAPIv1.6.43 drawv1.7 · 10 s rule57.399.470.7$0.0011*
  90. 100InstinctAPIv1.7 · 10 s rule47.864.470.5$0.015
  91. 101decider chat on Gemma-4-31B-itUsev1.7 · 10 s rule58.556.170.4$0.029*Use this model
  92. 102baselabs elephant-1Usev1.7.0 drawv1.7 · 10 s rule55.868.170.2$0.012*Use this model
  93. 103Jebadiah 27BUsev1.7 · 10 s rule57.557.570.2$0.026*Use this model
  94. 104GEV-26B-DecideUsev1.7 · 10 s rule49.872.670.1$0.0082*Use this model
  95. 105Chon-1.0-12B-FastUsev1.7.1 drawv1.7 · 10 s rule60.369.170.1$0.011*Use this model
  96. 106decisio v0.9.0 on gemma-4-12B-itUsev1.7 · 10 s rule52.367.770.0$0.012*Use this model
  97. 107Blink v0.3 26B-A4BUsev1.6.9 drawv1.7 · 10 s rule49.271.169.9$0.0092*Use this model
  98. 108NInfer Qwen3.8-Flash-Next mixedUsev1.7 · 10 s rule49.071.869.7$0.0087*Use this model
  99. 109attest-31B-v1.0APIv1.6.43 drawv1.7 · 10 s rule65.250.769.5$0.044*
  100. 110Respan Span-02Usev1.6.43 drawv1.7 · 10 s rule60.953.569.3$0.035*Use this model
  101. 111Philly Decider v2Usev1.7.10 drawv1.7 · 10 s rule56.179.169.3$0.0050*Use this model
  102. 112Wald 4B v2.1Usev1.6.9 drawv1.7 · 10 s rule52.060.869.3$0.020*Use this model
  103. 113JevOneUsev1.7 · 10 s rule46.072.868.9$0.0081*Use this model
  104. 114vLLM-SR d3-miniUsev1.7.0 drawv1.7 · 10 s rule50.878.868.8$0.0051*Use this model
  105. 115Torchcast Decision 4BUsev1.6.43 drawv1.7 · 10 s rule54.279.468.7$0.0049*Use this model
  106. 116iota-decide-4b-v0.3Usev1.7.10 drawv1.7 · 10 s rule52.578.968.6$0.0051*Use this model
  107. 117Xevn-4BUsev1.7.0 drawv1.7 · 10 s rule52.180.568.6$0.0045*Use this model
  108. 118Chon-1.0-E4BUsev1.7.3 drawv1.7 · 10 s rule55.583.568.5$0.0035*Use this model
  109. 119Hopper 12B trainedUsev1.7 · 10 s rule51.766.668.1$0.013*Use this model
  110. 120SPX-CD-OmniUsev1.7 · 10 s rule48.868.768.1$0.011*Use this model
  111. 121NInfer Qwen3.8-27B NVFP4Usev1.7 · 10 s rule51.856.568.0$0.028*Use this model
  112. 122Aplomb 1Usev1.7 · 10 s rule45.272.467.1$0.0083*Use this model
  113. 123Bespoke Nimble 9B v3Usev1.7 · 10 s rule56.565.866.8$0.014*Use this model
  114. 124Amazon Strands Decider 26B-A4BUsev1.7.1 drawv1.7 · 10 s rule46.771.966.7$0.0086*Use this model
  115. 125Tern-35B-A3BUsev1.7.0 drawv1.7 · 10 s rule44.171.366.4$0.0090*Use this model
  116. 126KnowLine 4B-Gen5.5Usev1.7.18 drawv1.7 · 10 s rule46.469.866.2$0.010*Use this model
  117. 127decider-35b-a3bUsev1.7 · 10 s rule48.773.266.2$0.0078*Use this model
  118. 128Metask rain 4BUsev1.6.4 drawv1.7 · 10 s rule51.169.566.0$0.010*Use this model
  119. 129Strike-1APIv1.7.16 drawv1.7 · 10 s rule42.554.166.0$0.034*
  120. 130shisa-de-1Usev1.6.43 drawv1.7 · 10 s rule49.971.165.5$0.0092*Use this model
  121. 131Plumb-4BUsev1.7 · 10 s rule42.877.665.3$0.0056*Use this model
  122. 132Arcbright Elf 1 12BUsev1.6.43 drawv1.7 · 10 s rule43.867.865.0$0.012*Use this model
  123. 133decider-4b v2Usev1.7 · 10 s rule40.179.164.9$0.0050*Use this model
  124. 134Seb-9B-v2Usev1.7.0 drawv1.7 · 10 s rule46.168.664.8$0.011*Use this model
  125. 135HS1 0.1.2APIv1.6.43 drawv1.7 · 10 s rule52.949.164.3$0.050*
  126. 136JevK5 v0.3Usev1.7 · 10 s rule38.577.664.2$0.0056*Use this model
  127. 137Saif 4BUsev1.8.1 drawv1.7 · 10 s rule40.181.264.1$0.0042*Use this model
  128. 138Bespoke Nimble-v4APIv1.6.43 drawv1.7 · 10 s rule41.460.764.0$0.020*
  129. 139Jeffan-4B-2010Usev1.7.0 drawv1.7 · 10 s rule40.178.764.0$0.0051*Use this model
  130. 140Clef-FlashUsev1.7 · 10 s rule42.266.563.8$0.013*Use this model
  131. 141JPT-4BUsev1.6.43 drawv1.7 · 10 s rule40.878.963.8$0.0051*Use this model
  132. 142Vansa-3.4APIv1.7 · 10 s rule40.961.963.7$0.019
  133. 143deck-4B v1.0Usev1.7 · 10 s rule38.678.063.6$0.0054*Use this model
  134. 144cribl-decision-1.0-9BUsev1.8.3 drawv1.7 · 10 s rule40.975.063.5$0.0068*Use this model
  135. 145Jebadiah 9B v2.1Usev1.6.43 drawv1.7 · 10 s rule37.268.563.4$0.011*Use this model
  136. 146GLiClass v4 9BUsev1.7.9 drawv1.7 · 10 s rule39.670.963.2$0.0093*Use this model
  137. 147Seb-9BUsev1.6.43 drawv1.7 · 10 s rule38.368.663.1$0.011*Use this model
  138. 148janus 4BUsev1.7 · 10 s rule37.679.062.9$0.0050*Use this model
  139. 149Amos 1 mediumAPIv1.6.43 drawv1.7 · 10 s rule38.148.762.9$0.051*
  140. 150JevK5 v0.2.0Usev1.7 · 10 s rule36.377.662.8$0.0056*Use this model
  141. 151cribl-decision-2Usev1.8.1 drawv1.7 · 10 s rule38.883.162.7$0.0037*Use this model
  142. 152Messier One v0.4Usev1.6.43 drawv1.7 · 10 s rule52.075.762.4$0.0065*Use this model
  143. 153Imajev-9BUsev1.6.43 drawv1.7 · 10 s rule36.750.562.4$0.045*Use this model
  144. 154Granite Switch 4.1 30BUsev1.8.5 drawv1.7 · 10 s rule44.646.662.1$0.060*Use this model
  145. 155Imajev-4BUsev1.7 · 10 s rule34.677.861.9$0.0055*Use this model
  146. 156jqvUsev1.7 · 10 s rule33.954.761.9$0.032*Use this model
  147. 157metask-jev-4bUsev1.7 · 10 s rule39.276.061.8$0.0063*Use this model
  148. 158Kahn1 4BUsev1.6.43 drawv1.7 · 10 s rule39.764.761.7$0.015*Use this model
  149. 159pplx-decider-flash-v1.2Usev1.6.43 drawv1.7 · 10 s rule45.579.061.7$0.0050*Use this model
  150. 160levUsev1.7 · 10 s rule41.176.361.7$0.0061*Use this model
  151. 161Decision 2.0 Lux 9BUsev1.6.43 drawv1.7 · 10 s rule35.868.161.6$0.012*Use this model
  152. 162JEB-35B-A3BUsev1.6.43 drawv1.7 · 10 s rule33.272.361.6$0.0084*Use this model
  153. 163Decisio v0.9.0 on Qwen3.6-35B-A3BUsev1.7.10 drawv1.7 · 10 s rule34.061.761.5$0.019*Use this model
  154. 164blink-4bUsev1.6.43 drawv1.7 · 10 s rule35.078.161.4$0.0054*Use this model
  155. 165HopperUsev1.7 · 10 s rule33.676.861.2$0.0059*Use this model
  156. 166Malkuth-4BUsev1.7 · 10 s rule33.877.561.1$0.0056*Use this model
  157. 167Qwen3.5-9B Jev-like data-mix v2Usev1.7 · 10 s rule41.965.460.9$0.014*Use this model
  158. 168Decision 4B v1.2Usev1.7 · 10 s rule32.477.660.4$0.0056*Use this model
  159. 169jiwo 4bUsev1.6.9 drawv1.7 · 10 s rule34.878.660.2$0.0052*Use this model
  160. 170Verdin-4BUsev1.7.0 drawv1.7 · 10 s rule35.476.759.6$0.0060*Use this model
  161. 171Wald 4BUsev1.6.4 drawv1.7 · 10 s rule41.070.559.6$0.0096*Use this model
  162. 172Quyet-1.0-MediumUsev1.7 · 10 s rule41.778.059.2$0.0054*Use this model
  163. 173Decision-4BUsev1.7 · 10 s rule37.980.459.1$0.0045*Use this model
  164. 174Jev-Style-2BUsev1.7.17 drawv1.7 · 10 s rule31.091.058.8$0.0020*Use this model
  165. 175Clef-omniUsev1.6.9 drawv1.7 · 10 s rule34.556.058.3$0.029*Use this model
  166. 176Manchego v2.1Usev1.7 · 10 s rule32.578.958.3$0.0051*Use this model
  167. 177Messier One v0.2Usev1.7 · 10 s rule42.479.458.1$0.0048Use this model
  168. 178Diffusion JevUsev1.7 · 10 s rule54.971.258.1$0.0091*Use this model
  169. 179jev-localUsev1.7 · 10 s rule41.978.557.5$0.0052*Use this model
  170. 180Instinct Dual 4BAPIv1.7 · 10 s rule28.375.057.2$0.0068
  171. 181NVIDIA Nemotron 3.5 Super VLAPIv1.8.7 drawv1.7 · 10 s rule57.547.257.2$0.057
  172. 182Decision 4B v1.1Usev1.7 · 10 s rule29.177.657.2$0.0056*Use this model
  173. 183JEV Qwen3.5-9B Base NVFP4Usev1.7 · 10 s rule29.347.657.1$0.056*Use this model
  174. 184iota-decide-4bUsev1.6.43 drawv1.7 · 10 s rule29.477.557.1$0.0056*Use this model
  175. 185Uniphore-4B-DeciderUsev1.6.43 drawv1.7 · 10 s rule34.279.157.1$0.0050*Use this model
  176. 186Jebadiah 9B v2Usev1.6.43 drawv1.7 · 10 s rule30.668.556.9$0.011*Use this model
  177. 187CoCo-Decision-4BUsev1.7 · 10 s rule32.779.756.9$0.0047*Use this model
  178. 188APUS-OpenJev-v1-9BUsev1.7 · 10 s rule47.268.456.9$0.011*Use this model
  179. 189Decider-4BUsev1.6.43 drawv1.7 · 10 s rule27.180.756.8$0.0044*Use this model
  180. 190typecastlmUsev1.7 · 10 s rule29.878.556.5$0.0052*Use this model
  181. 191Bespoke Nimble 9BUsev1.7 · 10 s rule43.156.456.5$0.028*Use this model
  182. 192pjev-2bUsev1.7.0 drawv1.7 · 10 s rule28.863.556.5$0.016*Use this model
  183. 193TypeCastLM 1.4.0Usev1.7 · 10 s rule35.479.356.4$0.0049*Use this model
  184. 194RYOTIDE-QwenUsev1.6.43 drawv1.7 · 10 s rule25.276.756.3$0.0060*Use this model
  185. 195Compass 0.2.0Usev1.6.43 drawv1.7 · 10 s rule28.972.156.1$0.0085*Use this model
  186. 196Standard One 3B SH v4Usev1.7.0 drawv1.7 · 10 s rule34.448.656.0$0.052*Use this model
  187. 197Standard One 3B SHUsev1.6.43 drawv1.7 · 10 s rule32.848.655.6$0.052*Use this model
  188. 198baselabs joey-1Usev1.8.1 drawv1.7 · 10 s rule23.589.155.4$0.0023*Use this model
  189. 199SemIfUsev1.7 · 10 s rule27.177.655.4$0.0056*Use this model
  190. 200ArseneLupin v1.1Usev1.6.43 drawv1.7 · 10 s rule28.075.255.1$0.0067*Use this model
  191. 201local-jev Qwen3.5-4BUsev1.7 · 10 s rule24.177.555.0$0.0056*Use this model
  192. 202Decision 2BUsev1.7 · 10 s rule22.966.454.8$0.013*Use this model
  193. 203Granite Switch 4.1 8BUsev1.8.4 drawv1.7 · 10 s rule29.359.454.8$0.023*Use this model
  194. 204Jez-large-1.0.0Usev1.7.0 drawv1.7 · 10 s rule25.683.354.4$0.0036*Use this model
  195. 205NomaUsev1.6.43 drawv1.7 · 10 s rule25.176.554.3$0.0061*Use this model
  196. 206spark-s1-4b-v6Usev1.7 · 10 s rule43.275.054.0$0.0068*Use this model
  197. 207Imajev-2BUsev1.6.43 drawv1.7 · 10 s rule21.267.354.0$0.012*Use this model
  198. 208Sors-2BUsev1.8.1 drawv1.7 · 10 s rule31.491.154.0$0.0020*Use this model
  199. 209Amber 4BUsev1.8.2 drawv1.7 · 10 s rule24.886.453.8$0.0028*Use this model
  200. 210RYOTIDE-Qwen9Usev1.6.4 drawv1.7 · 10 s rule27.666.152.5$0.013*Use this model
  201. 211classify-1APIv1.7.3 drawv1.7 · 10 s rule21.850.852.3$0.044
  202. 212Arbiter-4BUsev1.6.43 drawv1.7 · 10 s rule27.480.752.3$0.0044*Use this model
  203. 213DecisionNode Edge 0.9.2Usev1.7.0 drawv1.7 · 10 s rule26.886.852.1$0.0028*Use this model
  204. 214system-one-openAPIv1.7 · 10 s rule27.968.351.9$0.011*
  205. 215D1A on Gemma 4 E4BUsev1.7.7 drawv1.7 · 10 s rule19.981.251.1$0.0042*Use this model
  206. 216cribl-decision-2-miniUsev1.8.1 drawv1.7 · 10 s rule18.689.450.9$0.0022*Use this model
  207. 217ClassOne Qwen 3.5 9BUsev1.7 · 10 s rule42.269.050.5$0.011*Use this model
  208. 218kev 8BUsev1.7 · 10 s rule29.368.950.0$0.011*Use this model
  209. 219Liquid AI d1-3BUsev1.6.9 drawv1.7 · 10 s rule22.150.649.5$0.044*Use this model
  210. 220ZeroEntropy zerank-2Usev1.7 · 10 s rule15.660.649.4$0.021*Use this model
  211. 221JADEUsev1.7 · 10 s rule47.961.149.2$0.020*Use this model
  212. 222open-alternative-jevUsev1.7 · 10 s rule22.363.349.1$0.017*Use this model
  213. 223OpenSourceJevUsev1.7 · 10 s rule23.283.848.6$0.0035*Use this model
  214. 224Open-Jev 2BUsev1.7 · 10 s rule21.869.647.7$0.010*Use this model
  215. 225EXAONE-4.0-1.2B-JEV v0.3Usev1.7 · 10 s rule14.857.947.7$0.025*Use this model
  216. 226Malkuth-2BUsev1.7 · 10 s rule15.983.847.6$0.0035*Use this model
  217. 227DecisionNode Edge Mini 0.9.2Usev1.7.0 drawv1.7 · 10 s rule17.686.847.6$0.0028*Use this model
  218. 228Nemotron Diffusion 8BUsev1.7 · 10 s rule20.455.547.5$0.030*Use this model
  219. 229jiwo 0.8bUsev1.6.9 drawv1.7 · 10 s rule15.985.047.2$0.0032*Use this model
  220. 230decision-machine-1APIv1.7 · 10 s rule13.158.346.1$0.024
  221. 231decisor-4bUsev1.6.4 drawv1.7 · 10 s rule39.678.745.7$0.0051*Use this model
  222. 232D1A on Gemma 4 E2BUsev1.7.7 drawv1.7 · 10 s rule8.984.045.5$0.0034*Use this model
  223. 233Qwen3-Reranker-4BUsev1.7 · 10 s rule16.360.644.7$0.021*Use this model
  224. 234decider-2bUsev1.7 · 10 s rule20.785.844.6$0.0030*Use this model
  225. 235BAAI bge-reranker-v2-m3Usev1.7 · 10 s rule0.159.344.1$0.023*Use this model
  226. 236kev 4BUsev1.7 · 10 s rule19.577.944.0$0.0055*Use this model
  227. 237Mixedbread mxbai-rerank-base-v2Usev1.7 · 10 s rule1.160.343.9$0.021*Use this model
  228. 238lev-350mUsev1.7 · 10 s rule4.880.043.3$0.0046*Use this model
  229. 239Jez-small-1.0.0Usev1.7.0 drawv1.7 · 10 s rule4.686.243.2$0.0029*Use this model
  230. 240Certo v1Usev1.7 · 10 s rule0.296.943.2$0.0013*Use this model
  231. 241PurpleMIST-Mini-1.0Usev1.7.0 drawv1.7 · 10 s rule8.486.343.0$0.0029*Use this model
  232. 242Chon-1.0-12BUsev1.6.43 drawv1.7 · 10 s rule27.364.943.0$0.015*Use this model
  233. 243smalljev semantic-v9Usev1.7 · 10 s rule8.760.843.0$0.020*Use this model
  234. 244Vega-UUsev1.7.0 drawv1.7 · 10 s rule8.579.442.6$0.0048*Use this model
  235. 245jul fastUsev1.7 · 10 s rule7.569.042.6$0.011*Use this model
  236. 246Krino-2BUsev1.6.43 drawv1.7 · 10 s rule22.084.242.4$0.0034*Use this model
  237. 247Alibaba GTE Reranker ModernBERT-baseUsev1.7 · 10 s rule0.069.341.7$0.011*Use this model
  238. 248Jeetu Decider 2BUsev1.7.2 drawv1.7 · 10 s rule10.285.941.7$0.0030*Use this model
  239. 249watt-flash-0.1Usev1.7 · 10 s rule3.889.041.6$0.0023Use this model
  240. 250Decision FastUsev1.7 · 10 s rule6.186.640.6$0.0028*Use this model
  241. 251openJev Verdict 1.4Usev1.7 · 10 s rule5.086.640.4$0.0028*Use this model
  242. 252Quyet-1.0-Small-ENUsev1.7 · 10 s rule6.189.139.5$0.0023*Use this model
  243. 253ClassOne Gemma 4 E4BUsev1.6.9 drawv1.7 · 10 s rule28.282.138.4$0.0040*Use this model
  244. 254ClassOne Qwen 3.5 4BUsev1.6.9 drawv1.7 · 10 s rule27.379.538.1$0.0048*Use this model
  245. 255OpenDecisionUsev1.7 · 10 s rule7.479.138.0$0.0050*Use this model
  246. 256verdict-smallUsev1.7 · 10 s rule2.3100.037.6$0.00087*Use this model
  247. 257kev 0.6BUsev1.7 · 10 s rule7.684.236.9$0.0034*Use this model
  248. 258Vega 4BUsev1.6.9 drawv1.7 · 10 s rule15.761.736.6$0.019*Use this model
  249. 259WaterSheepUsev1.7 · 10 s rule5.368.735.5$0.011*Use this model
  250. 260Raw Phi-4 mini direct logitsUsev1.7 · 10 s rule13.753.235.3$0.036*Use this model
  251. 261Fastino GLiNER-2.5-DecideAPIv1.7 · 10 s rule7.045.834.6$0.064
  252. 262Raw Qwen3 4B Instruct 2507 direct logitsUsev1.7 · 10 s rule34.077.934.6$0.0055*Use this model
  253. 263Vega 0.8BUsev1.6.9 drawv1.7 · 10 s rule3.168.033.0$0.012*Use this model
  254. 264kev 0.5BUsev1.7 · 10 s rule4.684.232.9$0.0034*Use this model
  255. 265Liquid AI d1-omni-600MUsev1.6.9 drawv1.7 · 10 s rule3.373.132.8$0.0079*Use this model
  256. 266Quyet-1.0-SmallUsev1.7 · 10 s rule4.079.632.5$0.0048*Use this model
  257. 267Quyet-1.0-TinyUsev1.7 · 10 s rule3.788.431.1$0.0024*Use this model
  258. 268Open Jev JSON CanvasUsev1.7 · 10 s rule61.270.730.6$0.0095*Use this model
  259. 269Phocinae-Largha-150M-v1Usev1.6.43 drawv1.7 · 10 s rule3.695.330.2$0.0014*Use this model
  260. 270Raw Qwen3 8B direct logitsUsev1.7 · 10 s rule26.568.830.1$0.011*Use this model
  261. 271system-oneUsev1.7 · 10 s rule29.368.329.7$0.011*Use this model
  262. 272JevActAPIv1.7 · 10 s rule10.168.628.8$0.011*
  263. 273openJev VerdictUsev1.7 · 10 s rule4.186.626.6$0.0028*Use this model
  264. 274Tacet SonataUsev1.7 · 10 s rule2.289.024.2$0.0023*Use this model
  265. 275CLM-8BUsev1.7 · 10 s rule0.173.720.1$0.0075*Use this model
  266. 276ClassOne Qwen 3.5 2BUsev1.6.9 drawv1.7 · 10 s rule5.885.819.2$0.0030*Use this model
  267. 277Deem 0.8B v1Usev1.7 · 10 s rule6.586.818.1$0.0027*Use this model
  268. 278Laya multilingualUsev1.7 · 10 s rule0.382.217.0$0.0039*Use this model
  269. 279ClassOne Gemma 4 E2BUsev1.7 · 10 s rule7.885.815.2$0.0030*Use this model
  270. 280Julia-1Usev1.6.43 drawv1.7 · 10 s rule2.489.79.9$0.0022*Use this model
  271. 281Raw Qwen3 1.7B direct logitsUsev1.7 · 10 s rule5.184.18.3$0.0034*Use this model
  272. 282Raw Qwen3 0.6B direct logitsUsev1.7 · 10 s rule5.684.18.2$0.0034*Use this model

Wide coloured bar = Capability Score (0–100). Thin bars: cost above, median latency below; shared log ratio scale 0.25× … 64× Jev (v1.5 reference), ticks at 1× and 2× (cap). Shorter is cheaper or faster. * = est. (estimated cost). # counts ranked Jev-class systems; “–” marks unranked or outside systems. Tap ⓘ or hover a row or thin bar for details.

Cost and latency correlate here (Spearman ρ = 0.30, n = 319).

  • Jev — reference (TypeSafe, closed) (2)
  • Closed API (weights not public) (37)
  • Open weights · LLM decoder (240)
  • Open weights · diffusion LM (6)
  • Open weights · encoder / classifier (21)
  • Open weights · reranker (5)
  • Base model control (no decision fine-tune, raw logits) (5)
  • System (router / cascade / ensemble) (3)
  • green: ≤ reference
  • amber: ≤ cap (2× reference)
  • red: > cap

Use = run this open-weights model yourself: own endpoint, API, local install or hardware, on Decision Models. Decision Models is run by the same company as Benchmark Heaven. Every open-weights row gets the same link, whether or not we serve the model; it never affects scores or ranks. Independence

Show general-purpose LLMs and other systems outside the limits

Sorted by Capability Score, not numbered. Each row says which limit it misses, measured against Jev ($0.032 per 1,000 decisions, median 0.62 s).

  1. –GPT-6 Luna (medium)APIv1.7 · 10 s rule96.938.498.4$0.11Outside: cost 3.5× Jev (v1.5 reference), latency 3.1× Jev (v1.5 reference)
  2. –GPT-6 Luna (low)APIv1.7 · 10 s rule95.739.097.1$0.11Outside: cost 3.3× Jev (v1.5 reference), latency 3.1× Jev (v1.5 reference)
  3. –GPT-5.6 LunaAPIv1.7 · 10 s rule93.030.293.6$0.21Outside: cost 6.6× Jev (v1.5 reference), latency 2.6× Jev (v1.5 reference)
  4. –DeepSeek V4.1 FlashAPIv1.7 · 10 s rule84.519.689.6$0.48Outside: cost 14.9× Jev (v1.5 reference), latency 3.6× Jev (v1.5 reference)
  5. –AJev lm6qUsev1.7.0 drawv1.7 · 10 s rule71.544.281.4$0.073*Use this modelOutside: cost 2.2× Jev (v1.5 reference)
  6. –SPX-CD ProAPIv1.7 · 10 s rule64.643.178.5$0.079*Outside: cost 2.4× Jev (v1.5 reference)
  7. –Amos 1 largeAPIv1.6.43 drawv1.7 · 10 s rule68.230.078.0$0.21*Outside: cost 6.6× Jev (v1.5 reference)
  8. –Fastino GLiDEAPIv1.7 · 10 s rule66.135.977.0$0.14Outside: cost 4.3× Jev (v1.5 reference)
  9. –wity-1APIv1.7 · 10 s rule67.958.876.9$0.024Outside: latency 3.0× Jev (v1.5 reference)API price · eligibility checked at the developer's list price; at base-model pricing it would exceed the cost cap (2.70× Jev)
  10. –Aleria Judge 1.1.1Usev1.7.0 drawv1.7 · 10 s rule68.041.176.9$0.092*Use this modelOutside: cost 2.8× Jev (v1.5 reference)
  11. –OpenJevUsev1.7 · 10 s rule70.254.975.8$0.032*Use this modelOutside: latency 3.1× Jev (v1.5 reference)
  12. –SimpleJev Qwen3.8-27BUsev1.7 · 10 s rule57.942.574.9$0.083*Use this modelOutside: cost 2.6× Jev (v1.5 reference), latency 2.3× Jev (v1.5 reference)
  13. –Sage 1.4APIv1.6.43 drawv1.7 · 10 s rule63.759.573.3$0.022*Outside: latency 2.8× Jev (v1.5 reference)
  14. –Autoloops – Gemma 4 31B ITAPIv1.7 · 10 s rule58.140.572.7$0.096Outside: cost 3.0× Jev (v1.5 reference)
  15. –SimpleJev Qwen3.8-27BAPIv1.7 · 10 s rule52.914.971.6$0.69*Outside: cost 21.3× Jev (v1.5 reference), latency 2.1× Jev (v1.5 reference)
  16. –SimpleJev Qwen3.6-35B-A3BAPIv1.7 · 10 s rule56.335.168.3$0.15*Outside: cost 4.5× Jev (v1.5 reference), latency 2.1× Jev (v1.5 reference)
  17. –Open-Jev 27B v1.1Usev1.7 · 10 s rule52.541.768.0$0.088*Use this modelOutside: cost 2.7× Jev (v1.5 reference), latency 2.9× Jev (v1.5 reference)
  18. –LitJevUsev1.7 · 10 s rule43.355.766.6$0.030*Use this modelOutside: latency 5.1× Jev (v1.5 reference)
  19. –Bev / Bonsai 27BUsev1.7 · 10 s rule43.855.766.1$0.030*Use this modelOutside: latency 3.5× Jev (v1.5 reference)
  20. –Open-Jev 9BUsev1.7 · 10 s rule43.952.761.8$0.038*Use this modelOutside: latency 2.3× Jev (v1.5 reference)
  21. –Standard One 8B SHUsev1.6.9 drawv1.7 · 10 s rule34.543.260.0$0.078*Use this modelOutside: cost 2.4× Jev (v1.5 reference)
  22. –Gemini 3.1 Flash-LiteAPIv1.7 · 10 s rule58.629.259.8$0.23Outside: cost 7.1× Jev (v1.5 reference)
  23. –reflex 4BUsev1.7 · 10 s rule30.877.758.9$0.0055*Use this modelOutside: latency 4.7× Jev (v1.5 reference)
  24. –openjev-sglangAPIv1.7 · 10 s rule38.835.658.7$0.14*Outside: cost 4.3× Jev (v1.5 reference), latency 2.1× Jev (v1.5 reference)
  25. –Standard One 8BUsev1.7 · 10 s rule32.843.358.4$0.078*Use this modelOutside: cost 2.4× Jev (v1.5 reference)
  26. –Nev-2BUsev1.6.43 drawv1.6 · re-measure pending20.286.649.9$0.0028*Use this modelOutside: latency 20.8× Jev (v1.5 reference)
  27. –Qwen3.8 27BAPIv1.7 · 10 s rule33.40.048.6$2.18*Outside: cost 67.4× Jev (v1.5 reference), latency 13.5× Jev (v1.5 reference)
  28. –Laya typed-decisionsUsev1.7 · 10 s rule5.582.544.2$0.0038*Use this modelOutside: latency 3.5× Jev (v1.5 reference)
  29. –Qwen3.5-0.8B Decision ModelUsev1.6 · re-measure pending7.386.142.6$0.0029*Use this modelOutside: latency 18.3× Jev (v1.5 reference)
  30. –Gutsy 0.8B v0.3Usev1.6.9 drawv1.7 · 10 s rule8.684.942.2$0.0032*Use this modelOutside: latency 3.1× Jev (v1.5 reference)
  31. –Reflex Instinct 0.6BUsev1.6.43 drawv1.6 · re-measure pending4.086.438.6$0.0028*Use this modelOutside: latency 6.3× Jev (v1.5 reference)
  32. –Vega-UUsev1.6.43 drawv1.7 · 10 s rule5.385.836.1$0.0030*Use this modelOutside: latency 2.6× Jev (v1.5 reference)
  33. –open-jev-deberta-v3-largeUsev1.7 · 10 s rule2.877.635.4$0.0056*Use this modelOutside: latency 5.7× Jev (v1.5 reference)
  34. –modernbert-jev v2Usev1.6.9 drawv1.6 · re-measure pending2.670.434.6$0.0097*Use this modelOutside: latency 8.7× Jev (v1.5 reference)
  35. –LayaUsev1.7 · 10 s rule1.884.932.5$0.0032*Use this modelOutside: latency 3.3× Jev (v1.5 reference)
  36. –SimpleJevUsev1.7 · 10 s rule12.776.931.1$0.0059*Use this modelOutside: latency 14.9× Jev (v1.5 reference)
  37. –MirrorUsev1.7 · 10 s rule0.089.325.8$0.0023*Use this modelOutside: latency 2.1× Jev (v1.5 reference)

Jev-class = cost per decision at most 2× Jev's (≤ $0.065 per 1,000 decisions) and median latency at most 2× Jev's (≤ 1.23 s), the adjusted p50 — the same median the speed chart plots, not the four-axis Speed score. 282 of 319 systems qualify; the other 37, including the general-purpose LLMs, are listed below the divider in the ranking. The charts below show speed and cost beside Capability Score; the official JevBench Score weighs all four axes.

Capability against cost and speed

Jev-class systems are shown by default. Bubble size follows the official JevBench Score; official rank stays unchanged. The five most capable Jev-class systems are labelled.

Capability vs cost

Upper right is best: more capable and cheaper. Cost is USD per 1,000 decisions on a log scale. The dashed line is 2× Jev’s cost.

Capability vs cost: 282 systems. Upper right is best: more capable and cheaper.102030405060708090100$0.0010$0.010$0.10$ per 1,000 decisions (log)Capability↑2× Jev (v1.5 reference)← priciercheaper →1. Torchcast Decision 31B v1.02. MetaDecision-31B-Rain3. Karpo-Decision-31B4. kettei-fat5. Super Decider
282 systems. Tap a bubble for its values.

Capability vs speed

Upper right is best: more capable and faster. Speed here is the median-latency speed — the same adjusted median (p50) latency the Jev-class limit uses — a log scale, so each 20 points is 10× faster (median under the numbers). Self-hosted systems are plotted at the est. incl. network latency (2x measured + 150 ms); their measured loopback p50 (on-GPU, no network; not directly comparable with API latency) is in the tooltip. The dashed line is 2× Jev’s median latency.

Capability vs speed: 282 systems. Upper right is best: more capable and faster.010203040506070809010070≈3.2 s80≈1.0 s90≈316 ms100≈100 msMedian-latency speedCapability↑2× Jev (v1.5 reference)← slowerfaster →1. Torchcast Decision 31B v1.02. MetaDecision-31B-Rain3. Karpo-Decision-31B4. kettei-fat5. Super Decider
282 systems. Tap a bubble for its values.
  • Jev — reference (TypeSafe, closed) (2)
  • Closed API (weights not public) (37)
  • Open weights · LLM decoder (240)
  • Open weights · diffusion LM (6)
  • Open weights · encoder / classifier (21)
  • Open weights · reranker (5)
  • Base model control (no decision fine-tune, raw logits) (5)
  • System (router / cascade / ensemble) (3)
  • faint = outside Jev-class
357 of 357 systems

Model kind

Jev-class

Release

Parameters

No system in this release reports an exact parameter count.

Developer/API price $/1k
Base-model price $/1k
Scored cost $/1k
Alternative price $/1k
p50 latency (s)
p95 latency (s)

JevBench v1.8.11

JevBench Composite Score: 315 ranked systems

Official· four axes 0–100, equal-weight harmonic mean · Method notes ↓

Weights:
Adjust weights ↓

View by:Capability ↑

The official order weighs Intelligence, Calibration, Speed and Cost equally. Each button re-sorts the same systems by one axis.

Greener = stronger within its column.

319 of 319 systems, sorted by official rank, #1 first.

  1. 1Torchcast Decision 4B (v1.3)Usefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource80.5I 67C 88S 92K 79est.$0.0049
  2. 2decide-1.1APIClosed APIBase model: undisclosed*179.9I 73C 90S 91K 70$0.0098
  3. 3Torchcast Decision 31B v1.0Usefine-tune · 31BBase model: google/gemma-4-31B-itsource78.9I 85C 93S 87K 59est.$0.023
  4. 4Super DeciderAPIfine-tune · 26B-A4B · NVFP4Base model: google/diffusiongemma-26B-A4B-itmodel card*278.5I 79C 91S 85K 64$0.016
  5. 5Myles-G 12BAPIClosed APIBase model: google/gemma-4-12B-itmodel card*378.2I 57C 84S 85K 99est.$0.0011
  6. 6H2O-Lightning-4B v1.1Usefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource77.9I 60C 90S 92K 79est.$0.0052
  7. 7Arcbright Elf 1 12B (r2)Usefine-tune · 12B · bf16Base model: google/gemma-4-12B-itsource77.9I 64C 91S 93K 72est.$0.0088
  8. 8SD-JEV-4B-v12.1.0APIClosed APIBase model: undisclosed77.7I 63C 87S 82K 84est.$0.0035
  9. 9kettei-fatAPIClosed APIBase model: undisclosed77.7I 80C 93S 93K 57est.$0.026
  10. 10Wald 26B-A4BUsefine-tune · 26B-A4B · bf16Base model: google/gemma-4-26B-A4B-itsource77.5I 66C 88S 92K 71est.$0.0093
  11. 11Jeff-1.1-Qwen-4BUsefine-tune · 4B · BF16Base model: Qwen/Qwen3.5-4Bsource77.4I 65C 89S 78K 81est.$0.0044
  12. 12Karpo-Decision-31BUsefine-tune · 31B · bf16Base model: google/gemma-4-31B-itsource77.4I 80C 94S 88K 58est.$0.025
  13. 13z3-mini-soup-previewUsemerge · 4.5B · BF16Base model: Qwen/Qwen3.5-4Bsource77.3I 57C 87S 93K 83est.$0.0037
  14. 14Decisio 12B tuned commit-thinkUsemerge · 12B · BF16Base model: google/gemma-4-12B-itsource77.1I 66C 87S 91K 71est.$0.0095
  15. 15Torchcast Decision 4B (v1.2)Usefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource77.0I 60C 85S 91K 79est.$0.0049
  16. 16iota-decide-4b-v0.4Usefine-tune · 4B · BF16Base model: Qwen/Qwen3.5-4Bsource77.0I 57C 86S 92K 83est.$0.0037
  17. 17Fastino GLiDE (hosted API, 22 Oct 2026)APIClosed APIBase model: undisclosed*476.4I 76C 93S 87K 59$0.023
  18. 18d1-ultraAPIClosed APIBase model: undisclosed76.0I 70C 91S 89K 62est.$0.018
  19. 19Fastino GLiDE 1.2Usefine-tune · 27B · bf16Base model: undisclosed75.9I 76C 93S 85K 58est.$0.025
  20. 20Bobcat Flash 1.2Usefine-tune · 26B / 4B active · FP8 on loadBase model: google/gemma-4-26B-A4B-itsource75.8I 66C 81S 91K 71est.$0.0092
Show all 319 systems (295 more ranked, 4 more not ranked)
  1. 21SubQ Snap 1.1APIClosed APIBase model: undisclosed75.7I 76C 88S 85K 61est.$0.020
  2. 22H2O-Lightning-31B v1.0.7Usefine-tune · 31BBase model: google/gemma-4-31B-itsource75.7I 79C 89S 90K 56est.$0.029
  3. 23Microsoft-Decision-1 (updated weights, Oct 2026)APIClosed APIBase model: Qwen3.5-9Bmodel cardvendor statement*575.7I 64C 91S 89K 66$0.013
  4. 24StartLux-Decision v1.1 27B InstantUsefine-tune · 27BBase model: undisclosed75.6I 73C 92S 90K 58est.$0.025
  5. 25Drex MaxAPIClosed APIBase model: Qwen3.8-27Bvendor statement*675.4I 79C 90S 90K 55$0.031
  6. 26baselabs mammoth-1Usefine-tune · 27B · bf16Base model: Qwen/Qwen3.5-27Bsource75.4I 76C 94S 84K 57est.$0.027
  7. 27Decisio 12B tuned commitUsemerge · 12B · BF16Base model: google/gemma-4-12B-itsource75.2I 63C 89S 90K 67est.$0.013
  8. 28Jeff-1.1-MediumUsefine-tune · 12B · BF16Base model: google/gemma-4-12B-itsource75.1I 74C 89S 74K 66est.$0.013
  9. 29Philly Decider v2Usefine-tune · 4B · BF16Base model: h2oai/h2o-lightning-4bsource75.0I 56C 83S 93K 79est.$0.0050
  10. 30Perplexity Decider v1.1 27BUsefine-tune · 27BBase model: Qwen/Qwen3.8-27Bsource74.9I 76C 90S 86K 57est.$0.026
  11. 31Bespoke Nimble-v4-maxAPIClosed APIBase model: Qwen/Qwen3.8-27Bmodel card*774.8I 68C 91S 88K 61est.$0.020
  12. 32Chon-1.0-E4BUsefine-tune · 4BBase model: google/gemma-4-E4B-itsource74.7I 55C 82S 89K 84est.$0.0035
  13. 33vLLM-SR d3-turboUsefine-tune · 27BBase model: undisclosed74.7I 75C 90S 86K 57est.$0.027
  14. 34vLLM-SR d3Usefine-tune · 26BBase model: Qwen/Qwen3.8-27Bsource74.6I 73C 91S 87K 57est.$0.027
  15. 35Super Decider (thinking)APIfine-tune · 26B-A4B · NVFP4Base model: google/diffusiongemma-26B-A4B-itmodel card*874.5I 80C 92S 80K 56$0.028
  16. 36Decision-King v1.1Usefine-tune · 27B · BF16Base model: Qwen/Qwen3.8-27Bsource74.5I 73C 87S 83K 61est.$0.020
  17. 37KnowLine 26B-A4B-Gen2.5Usefine-tune · 26B-A4BBase model: google/gemma-4-26B-A4B-itsource74.4I 65C 89S 89K 62est.$0.018
  18. 38Torchcast Decision 4B (v1.1)Usefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource74.4I 54C 83S 94K 79est.$0.0049
  19. 39SubQ SnapAPIClosed APIBase model: undisclosed74.4I 72C 91S 86K 58est.$0.026
  20. 40KnowLine 35B-A3B-Gen1.5Usefine-tune · 35B-A3B · fp8Base model: Qwen/Qwen3.5-35B-A3Bsource74.1I 60C 93S 92K 64est.$0.016
  21. 41basal-omni-1.0-ultraUsemerge · 31BBase model: google/gemma-4-31B-itsource74.1I 74C 90S 87K 56est.$0.029
  22. 42Xor 26B-A4BUsemerge · 26B-A4BBase model: google/gemma-4-26B-A4B-itsource74.0I 57C 85S 91K 72est.$0.0085
  23. 43torchcast-decision-27bUsemerge · 27BBase model: startlux-models/StartLux-Decision-27Bsource 1source 2*974.0I 70C 90S 88K 58est.$0.026
  24. 44Arcbright Elf 1 31BUsefine-tune · 31B · bf16Base model: google/gemma-4-31B-itsource74.0I 72C 92S 90K 55est.$0.031
  25. 45pplx-decider-v1.2Usefine-tune · 27BBase model: Qwen/Qwen3.8-27Bsource73.9I 72C 90S 86K 57est.$0.026
  26. 46Piccolo 4BUsefine-tune · 4B · BF16Base model: Qwen/Qwen3.5-4Bsource73.9I 56C 86S 88K 74est.$0.0071
  27. 47Quyet-1.0-LargeUseadapter · 31BBase model: google/gemma-4-31B-itsource73.8I 73C 90S 86K 56est.$0.030
  28. 48Nimble-v4-maxUsefine-tune · 27BBase model: Qwen/Qwen3.8-27Bsource73.8I 71C 91S 85K 57est.$0.026
  29. 49Sage 1.3.0APIClosed APIBase model: undisclosed73.7I 66C 92S 92K 58est.$0.025
  30. 50decider-12b v2Usefine-tune · 12BBase model: google/gemma-4-12B-itsource73.7I 62C 81S 90K 68est.$0.012
  31. 51Xevn-4BUsefine-tune · 4B · bf16Base model: Qwen/Qwen3.5-4Bsource73.7I 52C 85S 92K 81est.$0.0045
  32. 52kettei-chubAPIClosed APIBase model: undisclosed73.4I 62C 84S 95K 63est.$0.018
  33. 53Surogate Rune 26B-A4B v3Use26B-A4BBase model: google/gemma-4-26B-A4B-itsource*1073.4I 56C 91S 87K 71est.$0.0092
  34. 54Jeff-1.0-LargeUsefine-tune · 31BBase model: google/gemma-4-31B-itsource73.4I 72C 91S 88K 55est.$0.033
  35. 55Lychee-27BUsefine-tune · 27B · bf16Base model: Qwen/Qwen3.8-27Bsource73.3I 74C 85S 84K 57est.$0.026
  36. 56oenogarumUsemerge · 31B · BF16Base model: google/gemma-4-31B-itsource73.3I 72C 91S 85K 56est.$0.030
  37. 57decisio v0.8.0 on gemma-4-31B-itUse31B · FP8 on loadBase model: google/gemma-4-31B-itsource73.2I 70C 89S 91K 55est.$0.032
  38. 58decider-12b v1, stock Gemma-4-12B-itUse12BBase model: google/gemma-4-12B-itsource73.2I 60C 83S 90K 68est.$0.012
  39. 59vLLM-SR d3-miniUsefine-tune · 4B · bf16Base model: Qwen/Qwen3.5-4Bsource73.2I 51C 87S 93K 79est.$0.0051
  40. 60Sieve-31B-v2Usefine-tune · 31B · BF16Base model: google/gemma-4-31B-itsource73.1I 69C 93S 86K 56est.$0.029
  41. 61MetaDecision-31B-RainUseadapter · 31B · W4A16 INT4Base model: google/gemma-4-31B-itsource73.0I 85C 92S 72K 55est.$0.032
  42. 62allec-oavgUsefine-tune · 31B · BF16Base model: google/gemma-4-31B-itsource73.0I 71C 91S 85K 56est.$0.030
  43. 63Piccolo 9BUsefine-tune · 9B · BF16Base model: Qwen/Qwen3.5-9Bsource72.8I 61C 88S 88K 64est.$0.016
  44. 64d1APIClosed APIBase model: undisclosed72.8I 62C 87S 88K 63$0.017
  45. 65Vera LargeUsefine-tune · 27B · bf16Base model: Qwen/Qwen3.8-27Bsource72.8I 68C 91S 88K 56est.$0.029
  46. 66iota-decide-4b-v0.3Usefine-tune · 4B · BF16Base model: Qwen/Qwen3.5-4Bsource72.8I 53C 85S 87K 79est.$0.0051
  47. 67Umpire 35B-A3BUsefine-tune · 35B / 3B active · bf16Base model: ornith-ai/Ornith-1.5-35B-A3Bsource72.5I 55C 89S 80K 75est.$0.0066
  48. 68swanOneUseNVFP4Base model: Mia-AiLab/Qwen3.8-Flash-Next-NVFP4source72.5I 56C 89S 83K 71est.$0.0090
  49. 69Piccolo 27BUsefine-tune · 27B · BF16Base model: Qwen/Qwen3.5-27Bsource72.4I 76C 91S 82K 53est.$0.037
  50. 70Mercury DecideAPISystemBase model: undisclosedsource72.2I 66C 81S 85K 62est.$0.018
  51. 71Winnow-12B Q8Usemerge · 12B · GGUF Q8_0Base model: google/gemma-4-12B-itsource72.2I 60C 83S 86K 67est.$0.013
  52. 72cribl-decision-1.0-27BUseadapter · 27B · BF16Base model: Qwen/Qwen3.5-27Bsource72.2I 59C 90S 86K 64est.$0.016
  53. 73Bobcat 1.3Usefine-tune · 31B · fp8Base model: google/gemma-4-31B-itsource72.2I 68C 88S 89K 55est.$0.031
  54. 74JPT-35B-A3BUsemerge · 35B-A3BBase model: Qwen/Qwen3.5-35B-A3Bsource72.2I 51C 90S 90K 72est.$0.0083
  55. 75Chon-1.0-12B-FastUseadapter · 12B · bf16Base model: google/gemma-4-12B-itsource72.0I 60C 80S 84K 69est.$0.011
  56. 76baselabs elephant-1Usefine-tune · 9B · bf16Base model: Qwen/Qwen3.5-9Bsource72.0I 56C 85S 90K 68est.$0.012
  57. 77metask-jev-rain-31B-trainedUseadapter · 31B · W4A16 INT4Base model: google/gemma-4-31B-itsource72.0I 71C 87S 82K 56est.$0.029
  58. 78deck-31BUse31B · FP8Base model: google/gemma-4-31B-itsource71.9I 73C 82S 86K 55est.$0.032
  59. 79CygnetUse12BBase model: google/gemma-4-12B-itsource71.9I 55C 87S 91K 67est.$0.013
  60. 80metask-jev-rain-12B-trainedUseadapter · 12B · W4A16 INT4Base model: google/gemma-4-12B-it-qat-q4_0-unquantizedsource71.7I 66C 87S 72K 65est.$0.014
  61. 81DOPEST 1.0 31BUsefine-tune · 31B · BF16Base model: google/gemma-4-31B-itsource71.7I 74C 90S 82K 53est.$0.038
  62. 82Eikos-27BUseadapter · 27BBase model: Qwen/Qwen3.8-27Bsource71.6I 64C 90S 88K 56est.$0.029
  63. 83Jev 1.13.0APIJev referenceBase model: undisclosed71.4I 64C 91S 91K 55$0.032
  64. 84Jeffan-31b-1710Usefine-tune · 31B · bf16Base model: google/gemma-4-31B-itsource71.3I 72C 92S 85K 51est.$0.042
  65. 85Kev 27BUsefine-tune · 27BBase model: Qwen/Qwen3.8-27Bsource71.3I 64C 94S 83K 56est.$0.029
  66. 86Gemini Decision 1APIClosed APIBase model: undisclosed*1171.3I 67C 90S 90K 53$0.038
  67. 87Jev-OmniUsefine-tuneBase model: google/gemma-4-12B-itsource71.0I 56C 87S 85K 66est.$0.013
  68. 88Celeris-1 DecisionAPIClosed APIBase model: undisclosed71.0I 59C 88S 87K 60est.$0.021
  69. 89GEV-26B-DecideUsedistilled · 26B-A4BBase model: google/gemma-4-26B-A4B-itsource70.8I 50C 90S 90K 73est.$0.0082
  70. 90decisio v0.8.0 on gemma-4-12B-itUse12BBase model: google/gemma-4-12B-itsource70.6I 52C 89S 86K 68est.$0.012
  71. 91Myles-G 31BAPIClosed APIBase model: google/gemma-4-31B-itmodel card*1270.6I 74C 84S 86K 51est.$0.043
  72. 92RSI-Jev v6.1-VL 27BUsemerge · 27BBase model: Qwen/Qwen3.8-27Bsource70.4I 62C 85S 85K 59est.$0.024
  73. 93Nagi-ENORMOUS 27BUseadapter · 27BBase model: Qwen/Qwen3.8-27Bsource70.3I 62C 85S 86K 57est.$0.026
  74. 94StartLux-Decision v1.1 27B AutoThinkUsefine-tune · 27BBase model: undisclosed70.2I 67C 91S 75K 57est.$0.028
  75. 95Jev 1.14.0APIJev referenceBase model: undisclosed70.2I 60C 90S 91K 55est.$0.032
  76. 96metask-jev-rain-31B-trained service v3Useadapter · 31B · W4A16 INT4Base model: google/gemma-4-31B-itsource70.2I 73C 87S 73K 55est.$0.031
  77. 97decisio v0.9.0 on gemma-4-12B-itUse12BBase model: google/gemma-4-12B-itsource70.1I 52C 88S 85K 68est.$0.012
  78. 98Bespoke Nimble 9B v3Useadapter · 9BBase model: Qwen/Qwen3.5-9Bsource70.0I 57C 77S 89K 66est.$0.014
  79. 99AutoTrust-Decider-Pro-27BUsedistilled · 27BBase model: Qwen/Qwen3.8-27Bsource69.9I 57C 91S 86K 59est.$0.024
  80. 100torchcast-decision-12bUseadapter · 12BBase model: google/gemma-4-12B-itsource69.8I 61C 83S 91K 56est.$0.028
  81. 101Messier One v0.4Usefine-tune · 4BBase model: undisclosed69.8I 52C 73S 89K 76est.$0.0065
  82. 102shisa-de-1Usefine-tune · 26B-A4BBase model: google/gemma-4-26B-A4B-itsource69.7I 50C 81S 93K 71est.$0.0092
  83. 103Microsoft-Decision-1 (Azure Foundry)APIClosed APIBase model: Qwen3.5-9Bvendor statementvendor statement*1369.1I 57C 84S 81K 61$0.019
  84. 104ClefUsefine-tuneBase model: Qwen/Qwen3.8-27BsourceWorkers AI price scenario (latency unmeasured): 29.4 (would be #193)69.0I 59C 92S 83K 55est.$0.030
  85. 105wity-1APIClosed APIBase model: undisclosedBase-model reference price (base undisclosed at the author's request): 43.0 (would be #164)69.0I 68C 86S 68K 59$0.024
  86. 106Wald 4B v2.1Usefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource68.9I 52C 87S 93K 61est.$0.020
  87. 107Hopper 12B trainedUsefine-tune · 12BBase model: google/gemma-4-12B-itsource68.8I 52C 85S 84K 67est.$0.013
  88. 108Blink v0.3 26B-A4BUsefine-tune · 26B-A4B · NVFP4Base model: google/gemma-4-26B-A4B-itsource68.7I 49C 91S 91K 71est.$0.0092
  89. 109Decision 2.0 Vega 27BUseadapter · 27BBase model: Qwen/Qwen3.8-27Bsource 1source 268.7I 60C 82S 83K 57est.$0.027
  90. 110René-1 31B FP8Usefine-tune · 31B · FP8Base model: google/gemma-4-31B-itsource68.6I 62C 90S 87K 51est.$0.042
  91. 111AutoJev-27BUsefine-tune · 27BBase model: Qwen/Qwen3.8-27Bsource68.6I 56C 88S 88K 57est.$0.028
  92. 112OpenJevUse26B-A4BBase model: google/diffusiongemma-26B-A4B-itsource 1source 268.4I 70C 81S 73K 55est.$0.032
  93. 113Jebadiah 27BUsemerge · 27BBase model: Qwen/Qwen3.8-27Bsource68.3I 57C 83S 85K 57est.$0.026
  94. 114decider chat on Gemma-4-31B-itUse31BBase model: google/gemma-4-31B-itsource68.1I 59C 82S 86K 56est.$0.029
  95. 115attest-31B-v1.0APIbase · 31BBase model: undisclosed67.9I 65C 74S 96K 51est.$0.044
  96. 116NInfer Qwen3.8-Flash-Next mixedUseLLM decoderBase model: Qwen3.8-Flash-Nextsource67.8I 49C 90S 89K 72est.$0.0087
  97. 117Metask rain 4BUsefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource67.7I 51C 81S 78K 70est.$0.010
  98. 118Sage 1.4APIClosed APIBase model: undisclosed67.7I 64C 83S 69K 59est.$0.022
  99. 119FrontiersMind Lumma-fev-27BUsefine-tune · 27B · bf16Base model: Qwen/Qwen3.8-27Bsource67.6I 55C 89S 85K 55est.$0.032
  100. 120Respan Span-02UseLLM decoderBase model: google/gemma-4-26B-A4B-itsource67.1I 61C 78S 86K 54est.$0.035
  101. 121NInfer Qwen3.8-27B NVFP4Use27BBase model: Qwen3.8-27Bsource66.7I 52C 84S 90K 56est.$0.028
  102. 122SPX-CD FlashAPISystemBase model: Qwen/Qwen3.6-35B-A3Bmodel card*1466.6I 58C 90S 79K 52est.$0.039
  103. 123decider-35b-a3bUsefine-tune · 35B-A3BBase model: Qwen/Qwen3.5-35B-A3B-Basesource66.5I 49C 84S 91K 73est.$0.0078
  104. 124Diffusion JevUse26B / 4B activeBase model: google/diffusiongemma-26B-A4B-itsource66.5I 55C 61S 86K 71est.$0.0091
  105. 125SPX-CD-OmniUsefine-tune · 12BBase model: google/gemma-4-12B-itsource66.0I 49C 87S 89K 69est.$0.011
  106. 126OpenAI DecisionsAPIClosed APIBase model: undisclosed62.4I 57C 90S 88K 49$0.052
  107. 127InstinctAPIClosed APIBase model: Qwen3.8-27Bsource*1562.2I 48C 93S 87K 64$0.015
  108. 128HS1 0.1.2APIClosed APIBase model: google/gemma-4-26B-A4B-it (self-reported)*1661.1I 53C 76S 94K 49est.$0.050
  109. 129KnowLine 31B-Gen1Usefine-tune · 31B · bf16Base model: google/gemma-4-31B-itsource60.2I 77C 93S 81K 46est.$0.061
  110. 130Amazon Strands Decider 26B-A4BUseadapter · 26B-A4B · bf16Base model: google/gemma-4-26B-A4B-itsource60.0I 47C 87S 88K 72est.$0.0086
  111. 131Turion Jev 31BUsefine-tune · 31B · fp8Base model: google/gemma-4-31B-itsource59.6I 71C 91S 86K 47est.$0.060
  112. 132KnowLine 4B-Gen5.5Usefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource59.2I 46C 86S 93K 70est.$0.010
  113. 133JevOneUseadapterBase model: Qwen/Qwen3.6-35B-A3Bsource58.7I 46C 92S 90K 73est.$0.0081
  114. 134Seb-9B-v2Usefine-tune · 9B · bf16Base model: Qwen/Qwen3.5-9Bsource57.6I 46C 84S 92K 69est.$0.011
  115. 135pplx-decider-flash-v1.2Usefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource56.7I 45C 78S 91K 79est.$0.0050
  116. 136APUS-OpenJev-v1-9BUsefine-tune · 9BBase model: Qwen/Qwen3.5-9Bsource56.6I 47C 67S 82K 68est.$0.011
  117. 137Aplomb 1Usefine-tune · 5.3BBase model: Qwen/Qwen3.5-4Bsource56.0I 45C 89S 89K 72est.$0.0083
  118. 138JADEUseadapter · 27BBase model: Qwen/Qwen3.8-27Bsource53.5I 48C 51S 85K 61est.$0.020
  119. 139Tern-35B-A3BUsefine-tune · 35B-A3B · bf16Base model: Qwen/Qwen3.6-35B-A3Bsource52.7I 44C 89S 91K 71est.$0.0090
  120. 140AJev lm6qUseadapter · 31B · bf16Base model: google/gemma-4-31B-itsource52.3I 71C 91S 82K 44est.$0.073
  121. 141NVIDIA Nemotron 3.5 Super VLAPIClosed APIBase model: Nemotron 3 Super Base (NVIDIA)model card*1752.2I 57C 57S 82K 47$0.057
  122. 142Arcbright Elf 1 12BUsefine-tune · 12B · bf16Base model: google/gemma-4-12B-itsource51.0I 44C 86S 92K 68est.$0.012
  123. 143Plumb-4BUsefine-tune · 4BBase model: alibiserikbay/JevK5source50.4I 43C 88S 94K 78est.$0.0056
  124. 144SPX-CD ProAPISystemBase model: Qwen/Qwen3.8-27Bmodel card*1847.6I 65C 92S 77K 43est.$0.079
  125. 145Messier One v0.2Usefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource47.4I 42C 74S 93K 79$0.0048
  126. 146spark-s1-4b-v6Useadapter · 4BBase model: Qwen3.5-4Bsource47.1I 43C 65S 86K 75est.$0.0068
  127. 147Bev / Bonsai 27BUse27B · GGUF PQ2_0Base model: Qwen/Qwen3.8-27Bsource46.6I 44C 89S 73K 56est.$0.030
  128. 148Clef-FlashUsefine-tuneBase model: Qwen/Qwen3.5-9BsourceWorkers AI price scenario (latency unmeasured): 39.1 (would be #176)46.1I 42C 85S 88K 66est.$0.013
  129. 149Quyet-1.0-MediumUseadapter · 4BBase model: Qwen/Qwen3.5-4Bsource45.7I 42C 77S 89K 78est.$0.0054
  130. 150Open-Jev 9BUseadapter · 9BBase model: Qwen/Qwen3.5-9Bsource45.5I 44C 80S 74K 53est.$0.038
  131. 151LitJevUse27BBase model: Qwen/Qwen3.8-27Bsource44.8I 43C 90S 68K 56est.$0.030
  132. 152Bespoke Nimble 9BUseadapter · 9BBase model: Qwen3.5-9Bsource44.5I 43C 70S 86K 56est.$0.028
  133. 153SimpleJev Qwen3.8-27B (self-hosted, v1.6 pool)Use27BBase model: Qwen/Qwen3.8-27Bsource44.5I 58C 92S 76K 42est.$0.083
  134. 154Qwen3.5-9B Jev-like data-mix v2Useadapter · 9BBase model: Qwen/Qwen3.5-9Bsource44.4I 42C 80S 85K 65est.$0.014
  135. 155cribl-decision-1.0-9BUseadapter · 9B · BF16Base model: Qwen/Qwen3.5-9Bsource44.4I 41C 86S 91K 75est.$0.0068
  136. 156levUseadapter · 4BBase model: Qwen/Qwen3.5-4Bsource44.3I 41C 82S 88K 76est.$0.0061
  137. 157Strike-1APIClosed APIBase model: undisclosed44.2I 42C 90S 83K 54est.$0.034
  138. 158JPT-4BUsemerge · 4BBase model: Qwen/Qwen3.5-4Bsource44.2I 41C 87S 88K 79est.$0.0051
  139. 159jev-localUse9BBase model: Qwen/Qwen3.5-9Bsource43.8I 42C 73S 73K 79est.$0.0052
  140. 160Aleria Judge 1.1.1Use31B · bf16Base model: undisclosed43.5I 68C 86S 86K 41est.$0.092
  141. 161Saif 4BUsefine-tune · 4B · BF16Base model: Qwen/Qwen3.5-4Bsource43.2I 40C 88S 91K 81est.$0.0042
  142. 162Bespoke Nimble-v4APIClosed APIBase model: undisclosed43.2I 41C 87S 89K 61est.$0.020
  143. 163decider-4b v2Useadapter · 4BBase model: Qwen/Qwen3.5-4B-Basesource43.1I 40C 90S 90K 79est.$0.0050
  144. 164Jeffan-4B-2010Usefine-tune · 4B · bf16Base model: Qwen/Qwen3.5-4Bsource43.0I 40C 88S 93K 79est.$0.0051
  145. 165ClassOne Qwen 3.5 9BUse9BBase model: Qwen/Qwen3.5-9Bsource43.0I 42C 59S 90K 69est.$0.011
  146. 166Vansa-3.4APIClosed APIBase model: Qwen3.5-4B (self-reported)*1942.6I 41C 87S 95K 62$0.019
  147. 167Wald 4BUsefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource42.4I 41C 78S 82K 71est.$0.0096
  148. 168Granite Switch 4.1 30BUseadapter · 30B · BF16Base model: ibm-granite/granite-4.1-30bsource40.8I 45C 80S 88K 47est.$0.060
  149. 169Open-Jev 27B v1.1Useadapter · 27BBase model: Qwen/Qwen3.8-27Bsource40.5I 52C 83S 72K 42est.$0.088
  150. 170GPT-6 Luna (low reasoning effort)APIClosed APIBase model: undisclosed40.4I 96C 98S 72K 39$0.108
  151. 171GLiClass v4 9BUsefine-tune · 9B · BF16Base model: Qwen/Qwen3.5-9Bsource40.2I 40C 87S 86K 71est.$0.0093
  152. 172Autoloops – Gemma 4 31B ITAPIClosed APIBase model: google/gemma-4-31B-itvendor statement*2040.2I 58C 87S 84K 40$0.096
  153. 173metask-jev-4bUse4BBase model: Qwen3.5-4Bsource39.9I 39C 84S 90K 76est.$0.0063
  154. 174cribl-decision-2Usefine-tune · 4B · BF16Base model: Qwen/Qwen3.5-4Bsource39.8I 39C 87S 91K 83est.$0.0037
  155. 175Kahn1 4BUsefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource39.6I 40C 84S 89K 65est.$0.015
  156. 176JevK5 v0.3UseadapterBase model: Qwen/Qwen3.5-4Bsource39.2I 39C 90S 94K 78est.$0.0056
  157. 177GPT-6 Luna (default medium reasoning effort)APIClosed APIBase model: undisclosed39.1I 97C 100S 71K 38$0.113
  158. 178deck-4B v1.0Useadapter · FP8Base model: alibiserikbay/JevK5 (built on Qwen/Qwen3.5-4B)source 1source 239.0I 39C 89S 90K 78est.$0.0054
  159. 179Seb-9BUsefine-tune · 9B · bf16Base model: Qwen/Qwen3.5-9Bsource37.2I 38C 88S 91K 69est.$0.011
  160. 180Decision-4BUseadapterBase model: Qwen/Qwen3.5-4Bsource37.0I 38C 80S 91K 80est.$0.0045
  161. 181decisor-4bUsefine-tune · 4B · FP8Base model: Qwen/Qwen3.5-4Bsource36.8I 40C 52S 92K 79est.$0.0051
  162. 182janus 4BUsefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource36.8I 38C 88S 93K 79est.$0.0050
  163. 183Jebadiah 9B v2.1Usemerge · 9BBase model: Qwen/Qwen3.5-9Bsource34.7I 37C 90S 91K 69est.$0.011
  164. 184JevK5 v0.2.0Useadapter · 4BBase model: Qwen3.5-4Bsource33.6I 36C 89S 91K 78est.$0.0056
  165. 185Amos 1 mediumAPI9BBase model: Qwen/Qwen3.5-9Bmodel card*2131.6I 38C 88S 86K 49est.$0.051
  166. 186Decision 2.0 Lux 9BUseLLM decoderBase model: vllm-sr/Decision-1.0-Lux-9Bsource31.5I 36C 87S 90K 68est.$0.012
  167. 187Verdin-4BUsemerge · 4BBase model: Qwen/Qwen3.5-4Bsource31.3I 35C 84S 90K 77est.$0.0060
  168. 188TypeCastLM 1.4.0Use3.8BBase model: Qwen/Qwen3.5-4Bsource31.1I 35C 77S 93K 79est.$0.0049
  169. 189blink-4bUsemerge · 4BBase model: Qwen/Qwen3.5-4Bsource30.8I 35C 88S 91K 78est.$0.0054
  170. 190Fastino GLiDEAPIClosed APIBase model: undisclosed30.6I 66C 88S 77K 36$0.137
  171. 191Imajev-9BUseadapter · 9BBase model: Qwen/Qwen3.5-9Bsource30.4I 37C 88S 79K 50est.$0.045
  172. 192jiwo 4bUsefine-tune · 4.2BBase model: Qwen/Qwen3.5-4Bsource30.3I 35C 86S 92K 79est.$0.0052
  173. 193Imajev-4BUseadapter · 4BBase model: Qwen/Qwen3.5-4Bsource30.0I 35C 89S 90K 78est.$0.0055
  174. 194Uniphore-4B-DeciderUsefine-tune · 4BBase model: undisclosed28.3I 34C 80S 87K 79est.$0.0050
  175. 195Malkuth-4BUseadapter · 4BBase model: Qwen/Qwen3.5-4B-Basesource28.2I 34C 88S 89K 77est.$0.0056
  176. 196HopperUseadapter · 4BBase model: Qwen/Qwen3.5-4Bsource27.7I 34C 89S 91K 77est.$0.0059
  177. 197SimpleJev Qwen3.6-35B-A3BAPI35B-A3BBase model: Qwen/Qwen3.6-35B-A3Bvendor statement*2227.5I 56C 80S 76K 35est.$0.145
  178. 198Decisio v0.9.0 on Qwen3.6-35B-A3BUsepipeline · 35B-A3B · FP8Base model: undisclosed27.2I 34C 89S 91K 62est.$0.019
  179. 199Clef-omniUsefine-tune · 30B-A3BBase model: Qwen/Qwen3-Omni-30B-A3B-Instructsource26.9I 34C 82S 87K 56est.$0.029
  180. 200JEB-35B-A3BUsefine-tune · 35B-A3BBase model: Qwen/Qwen3.6-35B-A3Bsource26.8I 33C 90S 93K 72est.$0.0084
  181. 201jqvUse32BBase model: Qwen3-32Bsource25.8I 34C 90S 82K 55est.$0.032
  182. 202CoCo-Decision-4BUsefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource25.7I 33C 81S 91K 80est.$0.0047
  183. 203Decision 4B v1.2Useadapter · 4BBase model: Qwen/Qwen3.5-4Bsource25.6I 32C 88S 94K 78est.$0.0056
  184. 204Manchego v2.1Useadapter · 4BBase model: Qwen/Qwen3.5-4Bsource25.5I 33C 84S 91K 79est.$0.0051
  185. 205Standard One 3B SH v4Usefine-tune · 3B · bf16Base model: StandardThinking/StandardOne-3Bsource24.4I 34C 78S 93K 49est.$0.052
  186. 206Sors-2BUsefine-tune · 2B · BF16Base model: Qwen/Qwen3.5-2B-Basesource23.6I 31C 77S 93K 91est.$0.0020
  187. 207Jev-Style-2BUsefine-tune · 2B · bf16Base model: Qwen/Qwen3.5-2Bsource23.6I 31C 87S 94K 91est.$0.0020
  188. 208Raw Qwen3 4B Instruct 2507 direct logitsUseoriginal · 4BBase model: Qwen/Qwen3-4B-Instruct-2507 (unmodified)source22.6I 34C 35S 91K 78est.$0.0055
  189. 209Standard One 3B SHUsefine-tune · 3BBase model: StandardThinking/StandardOne-3Bsource21.6I 33C 78S 86K 49est.$0.052
  190. 210GPT-5.6 LunaAPIClosed APIBase model: undisclosed21.4I 93C 94S 73K 30$0.213
  191. 211Jebadiah 9B v2Usemerge · 9BBase model: Qwen/Qwen3.5-9Bsource21.3I 31C 83S 91K 69est.$0.011
  192. 212reflex 4BUseadapter · 4BBase model: Qwen/Qwen3.5-4Bsource21.3I 31C 87S 69K 78est.$0.0055
  193. 213typecastlmUse4BBase model: Qwen3.5-4Bsource20.5I 30C 83S 93K 79est.$0.0052
  194. 214Amos 1 largeAPI27BBase model: Qwen/Qwen3.8-27Bauthor self-reportmodel card*2320.3I 68C 88S 83K 30est.$0.215
  195. 215iota-decide-4bUsefine-tune · 4B · bf16Base model: Qwen/Qwen3.5-4B-Basesource19.7I 29C 85S 87K 78est.$0.0056
  196. 216Decision 4B v1.1Useadapter · 4BBase model: Qwen/Qwen3.5-4Bsource19.5I 29C 85S 94K 78est.$0.0056
  197. 217Standard One 8B SHUsefine-tune · 8BBase model: StandardThinking/StandardOne-8Bsource18.7I 34C 85S 84K 43est.$0.078
  198. 218kev 8BUseadapter · 8BBase model: Qwen/Qwen3-8B-Basesource18.5I 29C 71S 88K 69est.$0.011
  199. 219Compass 0.2.0Useadapter · 4BBase model: Qwen/Qwen3.5-4Bsource18.5I 29C 83S 83K 72est.$0.0085
  200. 220Granite Switch 4.1 8BUseadapter · 8B · BF16Base model: ibm-granite/granite-4.1-8bsource18.5I 29C 80S 93K 59est.$0.023
  201. 221pjev-2bUsefine-tune · 2.5BBase model: openbmb/MiniCPM5-2Bsource18.1I 29C 84S 93K 64est.$0.016
  202. 222Instinct Dual 4BAPIClosed APIBase model: Qwen3.5-4Bsource*2418.0I 28C 86S 89K 75$0.0068
  203. 223ArseneLupin v1.1Useadapter · 4BBase model: Qwen/Qwen3.5-4Bsource17.3I 28C 82S 90K 75est.$0.0067
  204. 224Gemini 3.1 Flash-LiteAPIClosed APIBase model: Gemini 3 Pro (Google)model card*2516.9I 59C 61S 79K 29$0.230
  205. 225Standard One 8BUseadapter · 8BBase model: mistralai/Ministral-3-8B-Instruct-2512-BF16source16.9I 33C 84S 92K 43est.$0.078
  206. 226Arbiter-4BUsefine-tune · 4B · bf16Base model: Qwen/Qwen3.5-4Bsource16.5I 27C 77S 92K 81est.$0.0044
  207. 227system-one-openAPIadapter · 2BBase model: Gemma 4 E2Bsource16.4I 28C 76S 79K 68est.$0.011
  208. 228Decider-4BUsefine-tune · 4B · bf16Base model: Qwen/Qwen3.5-4Bsource16.3I 27C 87S 92K 81est.$0.0044
  209. 229SemIfUse4BBase model: Qwen/Qwen3.5-4Bsource16.1I 27C 84S 90K 78est.$0.0056
  210. 230RYOTIDE-Qwen9Usepipeline · 9B · BF16Base model: Qwen/Qwen3.5-9Bsource16.1I 28C 77S 86K 66est.$0.013
  211. 231ClassOne Gemma 4 E4BUsefine-tune · E4BBase model: google/gemma-4-E4B-itsource16.0I 28C 49S 92K 82est.$0.0040
  212. 232JEV Qwen3.5-9B Base NVFP4Use9B · NVFP4Base model: ig1/Qwen3.5-9B-NVFP4source16.0I 29C 85S 93K 48est.$0.056
  213. 233DecisionNode Edge 0.9.2Usefine-tune · 2B · bf16Base model: Qwen/Qwen3.5-2B-Basesource15.9I 27C 77S 93K 87est.$0.0028
  214. 234openjev-sglangAPI35B-A3BBase model: Qwen/Qwen3.6-35B-A3Bsource15.4I 39C 79S 78K 36est.$0.140
  215. 235ClassOne Qwen 3.5 4BUsefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource14.8I 27C 49S 91K 80est.$0.0048
  216. 236system-oneUse8BBase model: Qwen3-8Bsource14.7I 29C 30S 91K 68est.$0.011
  217. 237Jez-large-1.0.0Usefine-tune · E4B · bf16Base model: google/gemma-4-E4B-itsource14.2I 26C 83S 91K 83est.$0.0036
  218. 238Chon-1.0-12BUsefine-tune · 12BBase model: google/gemma-4-12B-itsource14.2I 27C 59S 65K 65est.$0.015
  219. 239RYOTIDE-QwenUsepipeline · 4BBase model: undisclosed13.5I 25C 87S 89K 77est.$0.0060
  220. 240NomaUseLLM decoderBase model: Qwen/Qwen3.5-4B-Basesource13.3I 25C 84S 93K 76est.$0.0061
  221. 241Amber 4BUseadapter · 4B · BF16Base model: Qwen/Qwen3.5-4Bsource13.1I 25C 83S 93K 86est.$0.0028
  222. 242Raw Qwen3 8B direct logitsUseoriginal · 8BBase model: Qwen/Qwen3-8B-Basesource12.1I 27C 34S 88K 69est.$0.011
  223. 243local-jev Qwen3.5-4BUsefine-tune · 4BBase model: Qwen/Qwen3.5-4Bsource12.0I 24C 86S 86K 78est.$0.0056
  224. 244baselabs joey-1Usefine-tune · 2B · BF16Base model: Qwen/Qwen3.5-2Bsource11.6I 23C 87S 92K 89est.$0.0023
  225. 245OpenSourceJevUseadapter · 4B · GGUF Q4_K_MBase model: Qwen3.5-4Bsource10.6I 23C 74S 76K 84est.$0.0035
  226. 246Decision 2BUseadapter · 2BBase model: openbmb/MiniCPM5-2Bsource10.3I 23C 87S 90K 66est.$0.013
  227. 247open-alternative-jevUse4BBase model: Qwen3.5-4Bsource9.4I 22C 76S 91K 63est.$0.017
  228. 248Krino-2BUseadapter · 2BBase model: Qwen/Qwen3.5-2B-Basesource9.2I 22C 63S 91K 84est.$0.0034
  229. 249Liquid AI d1-3BUsefine-tune · 3BBase model: LiquidAI/LFM2.5-VL-3Bsource8.9I 22C 77S 94K 51est.$0.044
  230. 250Open-Jev 2BUseadapter · 2BBase model: Qwen/Qwen3.5-2Bsource8.8I 22C 74S 77K 70est.$0.010
  231. 251classify-1APIClosed APIBase model: undisclosed8.5I 22C 83S 84K 51$0.044
  232. 252Imajev-2BUseadapter · 2BBase model: Qwen/Qwen3.5-2Bsource8.3I 21C 87S 81K 67est.$0.012
  233. 253decider-2bUseadapter · 2BBase model: Qwen/Qwen3.5-2B-Basesource8.0I 21C 69S 94K 86est.$0.0030
  234. 254D1A on Gemma 4 E4BUsefine-tune · BF16Base model: google/gemma-4-E4Bsource7.4I 20C 82S 92K 81est.$0.0042
  235. 255Nemotron Diffusion 8BUse8BBase model: nvidia/Nemotron-Labs-Diffusion-8Bsource7.3I 20C 75S 94K 55est.$0.030
  236. Nev-2B (re-measure pending under the 10 s rule)Usedistilled · 2B · GGUF Q8_0Base model: Qwen/Qwen3.5-2B-Basesource7.1I 20C 80S 55K 87est.$0.0028
  237. 256DeepSeek V4.1 FlashAPILLM decoderBase model: undisclosed6.9I 84C 95S 66K 20$0.480
  238. 257kev 4BUseadapter · 4BBase model: Qwen/Qwen3-4B-Basesource*266.8I 20C 68S 90K 78est.$0.0055
  239. 258cribl-decision-2-miniUsefine-tune · 0.8B · BF16Base model: Qwen/Qwen3.5-0.8Bsource6.3I 19C 83S 93K 89est.$0.0022
  240. 259DecisionNode Edge Mini 0.9.2Usefine-tune · 0.8B · bf16Base model: Qwen/Qwen3.5-0.8B-Basesource5.4I 18C 78S 93K 87est.$0.0028
  241. 260Malkuth-2BUseadapter · 2BBase model: empero-ai/Qwen3.8-2B-Distillsource4.1I 16C 79S 92K 84est.$0.0035
  242. 261Qwen3-Reranker-4BUseRerankerBase model: Qwen/Qwen3-4B-Basesource4.1I 16C 73S 81K 61est.$0.021
  243. 262jiwo 0.8bUsefine-tune · 0.75BBase model: Qwen/Qwen3.5-0.8Bsource4.1I 16C 79S 94K 85est.$0.0032
  244. 263ZeroEntropy zerank-2UseRerankerBase model: Qwen/Qwen3-4Bsource3.7I 16C 83S 82K 61est.$0.021
  245. 264Vega 4BUseadapter · 4BBase model: undisclosed3.6I 16C 57S 84K 62est.$0.019
  246. 265EXAONE-4.0-1.2B-JEV v0.3Use1.2BBase model: LGAI-EXAONE/EXAONE-4.0-1.2Bsource3.3I 15C 81S 94K 58est.$0.025
  247. 266SimpleJev Qwen3.8-27BAPI27BBase model: Qwen/Qwen3.8-27Bsource3.2I 53C 90S 76K 15est.$0.687
  248. 267Raw Phi-4 mini direct logitsUseoriginalBase model: microsoft/Phi-4-mini-instruct (unmodified)source2.5I 14C 57S 92K 53est.$0.036
  249. 268decision-machine-1APIClosed APIBase model: Qwen2.5-1.5B-Instruct (choices) + GLiNER2.5 (spans)vendor statement*272.4I 13C 79S 93K 58$0.024
  250. 269SimpleJevUse0.8BBase model: Qwen/Qwen3.5-0.8Bsource2.0I 13C 49S 59K 77est.$0.0059
  251. 270Jeetu Decider 2BUsefine-tune · 2B · bf16Base model: Qwen/Qwen3.5-2B-Basesource1.2I 10C 73S 91K 86est.$0.0030
  252. 271JevActAPIClosed APIBase model: Qwen3.5-2Bauthor self-report*281.1I 10C 47S 76K 69est.$0.011
  253. 272D1A on Gemma 4 E2BUsefine-tune · BF16Base model: google/gemma-4-E2Bsource0.8I 9C 82S 93K 84est.$0.0034
  254. 273smalljev semantic-v9Useadapter · 2BBase model: openbmb/MiniCPM5-2B-Basesource0.8I 9C 77S 90K 61est.$0.020
  255. 274Gutsy 0.8B v0.3Usefine-tune · 0.8B · GGUF Q8_0Base model: Qwen/Qwen3.5-0.8Bsource0.8I 9C 76S 72K 85est.$0.0032
  256. 275Vega-U (Qwen3.5-4B reader)Useadapter · 4BBase model: undisclosed0.7I 8C 77S 89K 79est.$0.0048
  257. 276PurpleMIST-Mini-1.0Usefine-tune · 2BBase model: Qwen/Qwen3.5-2B-Basesource0.7I 8C 78S 89K 86est.$0.0029
  258. 277kev 0.6BUseadapter · 0.6BBase model: Qwen/Qwen3-0.6B-Basesource0.5I 8C 66S 91K 84est.$0.0034
  259. 278jul fastUse2BBase model: openbmb/MiniCPM5-2Bsource0.5I 7C 78S 91K 69est.$0.011
  260. 279OpenDecisionUseEncoder / classifierBase model: MoritzLaurer/ModernBERT-large-zeroshot-v2.0source0.5I 7C 69S 86K 79est.$0.0050
  261. 280ClassOne Gemma 4 E2BUseE2BBase model: google/gemma-4-E2B-itsource0.5I 8C 23S 93K 86est.$0.0030
  262. Qwen3.5-0.8B Decision Model (re-measure pending under the 10 s rule)Usefine-tune · 0.8BBase model: Qwen/Qwen3.5-0.8B-Basesource0.5I 7C 78S 56K 86est.$0.0029
  263. 281Fastino GLiNER-2.5-DecideAPIfine-tune · 340MBase model: fastino/gliner2-large-v1source 1source 2*290.3I 7C 62S 85K 46$0.064
  264. 282Deem 0.8B v1UseLLM decoderBase model: Qwen/Qwen3.5-0.8Bsource0.3I 7C 30S 85K 87est.$0.0027
  265. 283Decision FastUseadapter · 0.6BBase model: Qwen/Qwen3-0.6B-Basesource0.3I 6C 75S 91K 87est.$0.0028
  266. 284Quyet-1.0-Small-ENUsefine-tune · 153MBase model: answerdotai/ModernBERT-basesource0.3I 6C 73S 95K 89est.$0.0023
  267. 285ClassOne Qwen 3.5 2BUsefine-tune · 2BBase model: Qwen/Qwen3.5-2Bsource0.2I 6C 33S 93K 86est.$0.0030
  268. 286Laya typed-decisionsUsefine-tune · 421MBase model: ModernBERT-largesource0.2I 5C 83S 69K 83est.$0.0038
  269. 287Vega-U (Qwen3.5-0.8B reader)Useadapter · 0.8BBase model: undisclosed0.2I 5C 67S 73K 86est.$0.0030
  270. 288WaterSheepUseEncoder / classifierBase model: answerdotai/ModernBERT-basesource0.2I 5C 66S 77K 69est.$0.011
  271. 289openJev Verdict 1.4UseEncoder / classifierBase model: knowledgator/gliclass-modern-base-v2.0source0.2I 5C 76S 81K 87est.$0.0028
  272. 290Raw Qwen3 0.6B direct logitsUseoriginal · 0.6BBase model: Qwen/Qwen3-0.6B-Basesource0.2I 6C 11S 93K 84est.$0.0034
  273. 291lev-350mUseadapterBase model: LiquidAI/LFM2.5-350Msource0.2I 5C 82S 94K 80est.$0.0046
  274. 292Raw Qwen3 1.7B direct logitsUseoriginal · 1.7BBase model: Qwen/Qwen3-1.7B-Basesource0.1I 5C 12S 93K 84est.$0.0034
  275. 293kev 0.5BUseadapter · 0.5BBase model: Qwen/Qwen2.5-0.5Bsource0.1I 5C 61S 92K 84est.$0.0034
  276. 294Jez-small-1.0.0Usefine-tune · 0.8B · fp32Base model: Qwen/Qwen3.5-0.8Bsource0.1I 5C 82S 93K 86est.$0.0029
  277. 295openJev VerdictUsefine-tuneBase model: knowledgator/gliclass-modern-base-v2.0source 1source 20.1I 4C 49S 78K 87est.$0.0028
  278. 296Quyet-1.0-SmallUsefine-tune · 328MBase model: aisingapore/SEA-LION-ModernBERT-300Msource0.1I 4C 61S 95K 80est.$0.0048
  279. Reflex Instinct 0.6B (re-measure pending under the 10 s rule)Useadapter · 0.6BBase model: Qwen/Qwen3-0.6Bsource0.1I 4C 73S 59K 86est.$0.0028
  280. 297watt-flash-0.1Usefine-tune · 140MBase model: jhu-clsp/mmBERT-smallsource0.1I 4C 79S 74K 89$0.0023
  281. 298Quyet-1.0-TinyUsedistilled · 183MBase model: jhu-clsp/mmBERT-smallsource0.1I 4C 58S 95K 88est.$0.0024
  282. 299Phocinae-Largha-150M-v1Usefine-tune · 144MBase model: undisclosed0.1I 4C 57S 77K 95est.$0.0014
  283. 300Liquid AI d1-omni-600MUsefine-tune · 600MBase model: LiquidAI/LFM2.5-Encoder-350Msource0.0I 3C 62S 94K 73est.$0.0079
  284. 301Vega 0.8BUseadapter · 0.8BBase model: Qwen/Qwen3.5-0.8Bsource0.0I 3C 63S 89K 68est.$0.012
  285. 302open-jev-deberta-v3-largeUseadapterBase model: microsoft/deberta-v3-largesource0.0I 3C 68S 67K 78est.$0.0056
  286. modernbert-jev v2 (re-measure pending under the 10 s rule)Usefine-tune · 150M · ONNX INT8Base model: undisclosed0.0I 3C 67S 58K 70est.$0.0097
  287. 303Julia-1Usefine-tune · 144MBase model: jhu-clsp/mmBERT-smallsource0.0I 2C 17S 78K 90est.$0.0022
  288. 304verdict-smallUsefine-tune · 118MBase model: intfloat/multilingual-e5-smallsource0.0I 2C 73S 89K 100est.$0.0009
  289. 305Tacet SonataUsefine-tune · 144MBase model: jhu-clsp/mmBERT-smallsource0.0I 2C 46S 79K 89est.$0.0023
  290. 306LayaUsefine-tune · 421MBase model: ModernBERT-largesource0.0I 2C 63S 73K 85est.$0.0032
  291. 307Mixedbread mxbai-rerank-base-v2UseRerankerBase model: Qwen2.5 (size not stated)source 1source 20.0I 1C 87S 92K 60est.$0.021
  292. 308Laya multilingualUsefine-tune · 322MBase model: mmBERT-basesource0.0I 0C 34S 77K 82est.$0.0039
  293. 309Certo v1UseEncoder / classifierBase model: ModernBERT-largesource0.0I 0C 86S 92K 97est.$0.0013
  294. 310CLM-8BUsefine-tune · 8BBase model: Qwen/Qwen3-8Bsource0.0I 0C 40S 92K 74est.$0.0075
  295. 311BAAI bge-reranker-v2-m3Usefine-tuneBase model: BAAI/bge-m3source0.0I 0C 88S 94K 59est.$0.023
  296. 312Alibaba GTE Reranker ModernBERT-baseUsefine-tuneBase model: answerdotai/ModernBERT-basesource0.0I 0C 83S 94K 69est.$0.011
  297. 313MirrorUseEncoder / classifierBase model: undisclosed0.0I 0C 52S 74K 89est.$0.0023
  298. 314Open Jev JSON CanvasUse26B-A4BBase model: google/diffusiongemma-26B-A4B-itsource0.0I 61C 0S 85K 71est.$0.0095
  299. 315Qwen3.8 27BAPI27BBase model: Qwen/Qwen3.8-27Bmodel card*300.0I 33C 64S 55K 0est.$2.178

Base-model notes

  1. *1 decide-1.1 (base model: undisclosed): No public base model is disclosed for this hosted API.
  2. *2 Super Decider (base model: google/diffusiongemma-26B-A4B-it): Vendor model card; the hosted endpoint was not inspected.
  3. *3 Myles-G 12B (base model: google/gemma-4-12B-it): Public model-card metadata; weights gated, deployed endpoint not inspected.
  4. *4 Fastino GLiDE (base model: undisclosed): No public base model declared; proprietary vendor-hosted API.
  5. *5 Microsoft-Decision-1 (base model: Qwen3.5-9B): Base model disclosed by Microsoft. The October weights update keeps the served model id and the model card's Version 1; Microsoft has announced a rebase only for future iterations.
  6. *6 Drex Max (base model: Qwen3.8-27B): Vendor's technical report. The vendor updated the model behind this API id after our 21 Oct measurement.
  7. *7 Bespoke Nimble-v4-max (base model: Qwen/Qwen3.8-27B): Vendor model card (published 18 Oct, after our 14 Oct measurement of the hosted API).
  8. *8 Super Decider (base model: google/diffusiongemma-26B-A4B-it): Vendor model card; the hosted endpoint was not inspected.
  9. *9 torchcast-decision-27b (base model: startlux-models/StartLux-Decision-27B): Direct base is StartLux-Decision-27B (startlux-models, CC-BY-NC-4.0), itself a modified Alibaba Qwen model; its card does not name the Qwen version. Qwen3.8-27B is inferred from the identical text_config.
  10. *10 Surogate Rune 26B-A4B v3 (base model: google/gemma-4-26B-A4B-it): This model also powers our System1 Models s1-pro service.
  11. *11 Gemini Decision 1 (base model: undisclosed): Google does not disclose a base model for this hosted API.
  12. *12 Myles-G 31B (base model: google/gemma-4-31B-it): Public model-card metadata; weights gated, deployed endpoint not inspected.
  13. *13 Microsoft-Decision-1 (base model: Qwen3.5-9B): Base model disclosed by Microsoft; the Decision-1 endpoint is evaluated as a hosted API offering.
  14. *14 SPX-CD Flash (base model: Qwen/Qwen3.6-35B-A3B): Vendor model card for the published adapter (9 Oct); the hosted Oct-4 checkpoint measured here is not stated to be identical.
  15. *15 Instinct (base model: Qwen3.8-27B): Operator-reported; weights not publicly verifiable.
  16. *16 HS1 0.1.2 (base model: google/gemma-4-26B-A4B-it (self-reported)): Customer statement (order notes, 21 Oct 2026): a LoRA fine-tune of Gemma-4-26B-A4B-it served in FP8. Not publicly documented and not independently verified.
  17. *17 NVIDIA Nemotron 3.5 Super VL (base model: Nemotron 3 Super Base (NVIDIA)): NVIDIA's own model.
  18. *18 SPX-CD Pro (base model: Qwen/Qwen3.8-27B): Vendor model card for the published adapter (9 Oct); the hosted Oct-4 checkpoint measured here is not stated to be identical.
  19. *19 Vansa-3.4 (base model: Qwen3.5-4B (self-reported)): Developer-reported to Benchmark Heaven (private correspondence, 1 Oct 2026): built on Qwen3.5-4B via an open decision fine-tune, with Vansa's own adapters; the intermediate fine-tune is not named. Not publicly documented or independently verified.
  20. *20 Autoloops – Gemma 4 31B IT (base model: google/gemma-4-31B-it): Vendor page.
  21. *21 Amos 1 medium (base model: Qwen/Qwen3.5-9B): Model card; the hosted endpoint was not inspected.
  22. *22 SimpleJev Qwen3.6-35B-A3B (base model: Qwen/Qwen3.6-35B-A3B): Vendor README.
  23. *23 Amos 1 large (base model: Qwen/Qwen3.8-27B): Vendor statement and model card; the hosted endpoint was not inspected.
  24. *24 Instinct Dual 4B (base model: Qwen3.5-4B): Operator-reported; weights not publicly verifiable.
  25. *25 Gemini 3.1 Flash-Lite (base model: Gemini 3 Pro (Google)): Google's own model.
  26. *26 kev 4B (base model: Qwen/Qwen3-4B-Base): Evaluated Qwen3 variant; later releases use a different base.
  27. *27 decision-machine-1 (base model: Qwen2.5-1.5B-Instruct (choices) + GLiNER2.5 (spans)): Vendor blog (19 Sep 2026); the vendor says the backend can change.
  28. *28 JevAct (base model: Qwen3.5-2B): Author's issue title only; not otherwise documented.
  29. *29 Fastino GLiNER-2.5-Decide (base model: fastino/gliner2-large-v1): Measured via Fastino's hosted API. A 340M English classifier (8k context), not built for multi-step reasoning: near chance on Noul and Score items; 23 very long items exceed its context. Our request follows Fastino's documented format. The open checkpoint scores 11.2 on the community Decision Index (rank 53 of 70; Jev 57.9). Fastino's larger GLiDE is a separate model, ranked separately on the API board.
  30. *30 Qwen3.8 27B (base model: Qwen/Qwen3.8-27B): The open model itself, not a fine-tune.
Weights:
Adjust weights ↓

Weights are relative: each axis counts in proportion to its slider. The score stays a weighted harmonic mean with the low-axis gates; the gates still apply when an axis sits at 0. Only equal weights give the official JevBench Score and rank.

Score = 4 / (1/I + 1/C + 1/S + 1/K) (each 0–100; × (axis / 50)² for Intelligence, Speed or Cost below 50)

  • Jev — reference (TypeSafe, closed) (2)
  • Closed API (weights not public) (37)
  • Open weights · LLM decoder (240)
  • Open weights · diffusion LM (6)
  • Open weights · encoder / classifier (21)
  • Open weights · reranker (5)
  • Base model control (no decision fine-tune, raw logits) (5)
  • System (router / cascade / ensemble) (3)
  • Striped bar = same system under the labelled alternative price assumption
  • Shown, not ranked
I, C, S, K = Intelligence, Calibration, Speed, Cost; the est. pill = estimated cost; ann. = announced price; API = the operator's endpoint saw held-out benchmark inputs, without answers; $/1k decisions = US dollars per 1,000 decisions (not heat-shaded). Names link to each project.

Compare two systems

Pick any two. Radars for the score axes, capability by subject topic and use cases, chance-corrected competence per request type on the open and sealed sets, and competence per tier on each set. Further out is better on every spoke; the link keeps the pair.

  • A: Jev 1.13.0REFERENCE · TypeSafe — Jev — reference (TypeSafe, closed) · Score 71.4 (#83)
  • B: Torchcast Decision 4B — Open weights · LLM decoder · Score 80.5 (#1)

The four score axes

Radar: the four score axes, two systemsThe four score axes, Jev 1.13.0 vs Torchcast Decision 4B. Intelligence: 63.6 vs 67.1; Calibration: 90.6 vs 87.8; Speed: 91.2 vs 92.4; Cost: 54.7 vs 79.4.50100Intelligence: Jev 1.13.0: 63.6; Torchcast Decision 4B: 67.1Intelligence63.6 · 67.1Calibration: Jev 1.13.0: 90.6; Torchcast Decision 4B: 87.8Calibration90.6 · 87.8Speed: Jev 1.13.0: 91.2; Torchcast Decision 4B: 92.4Speed91.2 · 92.4Cost: Jev 1.13.0: 54.7; Torchcast Decision 4B: 79.4Cost54.7 · 79.4
0–100, the values in the table. An axis a system has no published value for is left as a gap (it counts as 0 in the composite).

Capability by subject topic

Torchcast Decision 4B: Subject topics and use cases come from this draw's own item labels, added after its release with the same labelling recipe as the earlier draws (hand check 71/75).

Radar: capability by subject topic, two systemsCapability by subject topic, Jev 1.13.0 vs Torchcast Decision 4B. Rules, policy & law: 49.4 vs 73.0; Coding & software: 64.3 vs 70.0; Math & numbers: 10.4 vs 41.6; Finance & commerce: 40.9 vs 78.0; Support & operations: 48.8 vs 76.5; Everyday language: 77.5 vs 45.2; Safety & security: 52.2 vs 72.8. Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.−1000100Rules, policy & law: Jev 1.13.0: 49.4; Torchcast Decision 4B: 73.0 — Rules, policy & law: applying written rules: company policies, contracts, regulations, eligibility. 1616 items (151 open, 1465 sealed).Rules, policy& law49.4 · 73.0Coding & software: Jev 1.13.0: 64.3; Torchcast Decision 4B: 70.0 — Coding & software: code, SQL, repositories, developer tools and IT systems. 382 items (62 open, 320 sealed).Coding &software64.3 · 70.0Math & numbers: Jev 1.13.0: 10.4; Torchcast Decision 4B: 41.6 — Math & numbers: a calculation decides the answer: arithmetic, word problems, probability, dates, units. 370 items (37 open, 333 sealed).Math &numbers10.4 · 41.6Finance & commerce: Jev 1.13.0: 40.9; Torchcast Decision 4B: 78.0 — Finance & commerce: money: payments, refunds, invoices, orders, expenses, insurance payouts. 223 items (12 open, 211 sealed).Finance &commerce40.9 · 78.0Support & operations: Jev 1.13.0: 48.8; Torchcast Decision 4B: 76.5 — Support & operations: support tickets, incidents, logistics, scheduling desks and routing work to a team. 160 items (21 open, 139 sealed).Support &operations48.8 · 76.5Everyday language: Jev 1.13.0: 77.5; Torchcast Decision 4B: 45.2 — Everyday language: short everyday messages: intents, assistant requests, reading a detail out of a text. 145 items (11 open, 134 sealed).Everydaylanguage77.5 · 45.2Safety & security: Jev 1.13.0: 52.2; Torchcast Decision 4B: 72.8 — Safety & security: untrusted or injected instructions, fraud, moderation, access and security triage. 109 items (6 open, 103 sealed).Safety &security52.2 · 72.8

Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.

Measured subject-topic competence on this release’s item set. Chance-corrected competence per category (0 = at or below chance; negative averages are reported as 0, 100 = perfect), Jev 1.13.0: S+P+L1+L2+L3; Torchcast Decision 4B: S+P (draw v1.6-regular-20261012) items pooled; Breakdown values are shown for cells with at least 30 scored items; more items are being added.. Hover a category for its definition and item count.
What each category means · items per category
  • Rules, policy & law — applying written rules: company policies, contracts, regulations, eligibility. 1616 items (151 open / 1465 sealed)
  • Coding & software — code, SQL, repositories, developer tools and IT systems. 382 items (62 open / 320 sealed)
  • Math & numbers — a calculation decides the answer: arithmetic, word problems, probability, dates, units. 370 items (37 open / 333 sealed)
  • Finance & commerce — money: payments, refunds, invoices, orders, expenses, insurance payouts. 223 items (12 open / 211 sealed)
  • Support & operations — support tickets, incidents, logistics, scheduling desks and routing work to a team. 160 items (21 open / 139 sealed)
  • Everyday language — short everyday messages: intents, assistant requests, reading a detail out of a text. 145 items (11 open / 134 sealed)
  • Safety & security — untrusted or injected instructions, fraud, moderation, access and security triage. 109 items (6 open / 103 sealed)

Use cases (TypeSafe categories)

Torchcast Decision 4B: Subject topics and use cases come from this draw's own item labels, added after its release with the same labelling recipe as the earlier draws (hand check 71/75).

Radar: use cases (typesafe categories), two systemsUse cases (TypeSafe categories), Jev 1.13.0 vs Torchcast Decision 4B. Model routing: 73.0 vs 76.2; Legal & compliance: 68.6 vs 74.8; Customer support: 49.6 vs 65.6; Other: 40.1 vs 51.1; E-commerce: 40.5 vs 66.6; Insurance claims: 46.1 vs —; Risk assessment: 32.8 vs 54.7; Financial crime: 35.8 vs 36.1; Feature extraction: 28.3 vs —; Lead generation: 49.2 vs 66.9; Recruiting: 0.0 vs —; Knowledge graphs: 28.8 vs —; LLM guardrails: 68.0 vs —; Moderation: 51.6 vs 74.9; Code linting: 29.1 vs —; Search & retrieval: 80.9 vs —; Science: 21.2 vs —; Advertising: 16.4 vs —; Gaming: 5.1 vs —; Demand forecasting: 0.0 vs —. Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.−1000100Model routing: Jev 1.13.0: 73.0; Torchcast Decision 4B: 76.2 — Model routing: deciding which LLM or model tier should handle a prompt; classifying a request's intent, domain, difficulty or risk to pick a model. 524 items (86 open, 438 sealed).Modelrouting73.0 · 76.2Legal & compliance: Jev 1.13.0: 68.6; Torchcast Decision 4B: 74.8 — Legal and compliance: contracts, policies, regulations, eligibility rules, checking documents against written requirements. 515 items (74 open, 441 sealed).Legal &compliance68.6 · 74.8Customer support: Jev 1.13.0: 49.6; Torchcast Decision 4B: 65.6 — Customer support: classifying tickets and customer intents, urgency, refunds, routing cases to a team, checking support replies. 240 items (37 open, 203 sealed).Customersupport49.6 · 65.6Other: Jev 1.13.0: 40.1; Torchcast Decision 4B: 51.1 — Other: none of the above: general reasoning, arithmetic or date puzzles, everyday assistant requests without a business workflow. 174 items (24 open, 150 sealed).Other40.1 · 51.1E-commerce: Jev 1.13.0: 40.5; Torchcast Decision 4B: 66.6 — E-commerce marketplaces: product listings, product attributes, orders and deliveries, seller catalog policy. 119 items (9 open, 110 sealed).E-commerce40.5 · 66.6Insurance claims: Jev 1.13.0: 46.1; Torchcast Decision 4B: No measured result — Insurance claims: first-notice-of-loss reports, claim complexity, missing information, claim triage. 119 items (9 open, 110 sealed).Insuranceclaims46.1 · n/aRisk assessment: Jev 1.13.0: 32.8; Torchcast Decision 4B: 54.7 — Risk assessment: estimating probabilities and severity of risks, incidents, vendor or underwriting risk. 118 items (8 open, 110 sealed).Riskassessment32.8 · 54.7Financial crime: Jev 1.13.0: 35.8; Torchcast Decision 4B: 36.1 — Financial crime: suspicious transactions, KYC, fraud alerts, entity matching for investigations. 101 items (9 open, 92 sealed).Financialcrime35.8 · 36.1Feature extraction: Jev 1.13.0: 28.3; Torchcast Decision 4B: No measured result — Feature extraction for predictive modeling: turning natural-language data into probabilistic features or estimates for a downstream prediction. 99 items (5 open, 94 sealed).Featureextraction28.3 · n/aLead generation: Jev 1.13.0: 49.2; Torchcast Decision 4B: 66.9 — Lead generation: matching companies or inbound messages to an ideal customer profile, buyer intent, prioritising and routing sales leads. 95 items (4 open, 91 sealed).Leadgeneration49.2 · 66.9Recruiting: Jev 1.13.0: 0.0; Torchcast Decision 4B: No measured result — Recruiting: resumes, applications, interview feedback, matching candidates to roles. 94 items (1 open, 93 sealed).Recruiting0.0 · n/aKnowledge graphs: Jev 1.13.0: 28.8; Torchcast Decision 4B: No measured result — Graphs and knowledge graphs: entity types, relationships between records, contradictions between facts, multi-step lookups across linked facts. 94 items (3 open, 91 sealed).Knowledgegraphs28.8 · n/aLLM guardrails: Jev 1.13.0: 68.0; Torchcast Decision 4B: No measured result — LLM guardrails: checking LLM inputs, outputs and tool calls: prompt injection, jailbreaks, policy violations, tool-call errors, answer quality. 93 items (5 open, 88 sealed).LLMguardrails68.0 · n/aModeration: Jev 1.13.0: 51.6; Torchcast Decision 4B: 74.9 — Moderation and trust and safety: user content moderation: toxicity, harassment, spam, unsafe advice, personal data exposure. 93 items (7 open, 86 sealed).Moderation51.6 · 74.9Code linting: Jev 1.13.0: 29.1; Torchcast Decision 4B: No measured result — Semantic code linting: checking code or writing against conventions and guidelines, as in CI review of code. 90 items (3 open, 87 sealed).Code linting29.1 · n/aSearch & retrieval: Jev 1.13.0: 80.9; Torchcast Decision 4B: No measured result — Search and retrieval: scoring query-to-candidate relevance, reranking results, selecting useful context for RAG. 90 items (5 open, 85 sealed).Search &retrieval80.9 · n/aScience: Jev 1.13.0: 21.2; Torchcast Decision 4B: No measured result — Scientific discovery: screening papers, labelling research passages or survey answers, checking that citations support claims, methodology checks. 90 items (2 open, 88 sealed).Science21.2 · n/aAdvertising: Jev 1.13.0: 16.4; Torchcast Decision 4B: No measured result — Advertising: ad creatives, campaign copy, brand safety, prohibited claims in ads. 88 items (3 open, 85 sealed).Advertising16.4 · n/aGaming: Jev 1.13.0: 5.1; Torchcast Decision 4B: No measured result — Gaming: player reports, in-game chat, game support. 86 items (3 open, 83 sealed).Gaming5.1 · n/aDemand forecasting: Jev 1.13.0: 0.0; Torchcast Decision 4B: No measured result — Demand forecasting: purchase intent, product interest, demand and supply signals for forecasting. 83 items (3 open, 80 sealed).Demandforecasting0.0 · n/a

Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.

Measured application-category competence on this release’s item set. Chance-corrected competence per category (0 = at or below chance; negative averages are reported as 0, 100 = perfect), Jev 1.13.0: S+P+L1+L2+L3; Torchcast Decision 4B: S+P (draw v1.6-regular-20261012) items pooled; Breakdown values are shown for cells with at least 30 scored items; more items are being added.. Hover a category for its definition and item count.B (Torchcast Decision 4B) is measured on 9 of 20 spokes only, so it is drawn as points, not a shape — measured on fewer items; full radar after the next re-measure.Gaps, not zeros: Open spokes (n/a or n=…) have fewer than 30 answered items for that system or no published value.A point on the bold 0 ring printed 0.0 is a measured value, not a gap: these published cells clip below-chance results to 0, so 0 means at or below chance. Their score markers cannot fall inside that ring.
What each category means · items per category
  • Model routing — deciding which LLM or model tier should handle a prompt; classifying a request's intent, domain, difficulty or risk to pick a model. 524 items (86 open / 438 sealed)
  • Legal and compliance — contracts, policies, regulations, eligibility rules, checking documents against written requirements. 515 items (74 open / 441 sealed)
  • Customer support — classifying tickets and customer intents, urgency, refunds, routing cases to a team, checking support replies. 240 items (37 open / 203 sealed)
  • Other — none of the above: general reasoning, arithmetic or date puzzles, everyday assistant requests without a business workflow. 174 items (24 open / 150 sealed)
  • E-commerce marketplaces — product listings, product attributes, orders and deliveries, seller catalog policy. 119 items (9 open / 110 sealed)
  • Insurance claims — first-notice-of-loss reports, claim complexity, missing information, claim triage. 119 items (9 open / 110 sealed)
  • Risk assessment — estimating probabilities and severity of risks, incidents, vendor or underwriting risk. 118 items (8 open / 110 sealed)
  • Financial crime — suspicious transactions, KYC, fraud alerts, entity matching for investigations. 101 items (9 open / 92 sealed)
  • Feature extraction for predictive modeling — turning natural-language data into probabilistic features or estimates for a downstream prediction. 99 items (5 open / 94 sealed)
  • Lead generation — matching companies or inbound messages to an ideal customer profile, buyer intent, prioritising and routing sales leads. 95 items (4 open / 91 sealed)
  • Recruiting — resumes, applications, interview feedback, matching candidates to roles. 94 items (1 open / 93 sealed)
  • Graphs and knowledge graphs — entity types, relationships between records, contradictions between facts, multi-step lookups across linked facts. 94 items (3 open / 91 sealed)
  • LLM guardrails — checking LLM inputs, outputs and tool calls: prompt injection, jailbreaks, policy violations, tool-call errors, answer quality. 93 items (5 open / 88 sealed)
  • Moderation and trust and safety — user content moderation: toxicity, harassment, spam, unsafe advice, personal data exposure. 93 items (7 open / 86 sealed)
  • Semantic code linting — checking code or writing against conventions and guidelines, as in CI review of code. 90 items (3 open / 87 sealed)
  • Search and retrieval — scoring query-to-candidate relevance, reranking results, selecting useful context for RAG. 90 items (5 open / 85 sealed)
  • Scientific discovery — screening papers, labelling research passages or survey answers, checking that citations support claims, methodology checks. 90 items (2 open / 88 sealed)
  • Advertising — ad creatives, campaign copy, brand safety, prohibited claims in ads. 88 items (3 open / 85 sealed)
  • Gaming — player reports, in-game chat, game support. 86 items (3 open / 83 sealed)
  • Demand forecasting — purchase intent, product interest, demand and supply signals for forecasting. 83 items (3 open / 80 sealed)

Competence per request type, open / sealed

Radar: competence per request type, open / sealed, two systemsCompetence per request type, open / sealed, Jev 1.13.0 vs Torchcast Decision 4B. Choice · open: 79.0 vs 81.7; Choice · sealed: 75.8 vs 73.0; Noul · open: 54.5 vs 80.1; Noul · sealed: 46.1 vs 70.9; Score · open: 63.3 vs 47.5; Score · sealed: 63.0 vs 49.8. Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.−1000100Choice · open: Jev 1.13.0: 79.0; Torchcast Decision 4B: 81.7Choice · open79.0 · 81.7Choice · sealed: Jev 1.13.0: 75.8; Torchcast Decision 4B: 73.0Choice ·sealed75.8 · 73.0Noul · open: Jev 1.13.0: 54.5; Torchcast Decision 4B: 80.1Noul · open54.5 · 80.1Noul · sealed: Jev 1.13.0: 46.1; Torchcast Decision 4B: 70.9Noul · sealed46.1 · 70.9Score · open: Jev 1.13.0: 63.3; Torchcast Decision 4B: 47.5Score · open63.3 · 47.5Score · sealed: Jev 1.13.0: 63.0; Torchcast Decision 4B: 49.8Score ·sealed63.0 · 49.8

Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.

Chance-corrected competence for Choice, Noul and Score on the 300 open and 1200 sealed decisions.

Competence per tier — open set

Radar: competence per tier — open set, two systemsCompetence per tier — open set, Jev 1.13.0 vs Torchcast Decision 4B. Easy: 93.9 vs 86.2; Standard: 63.2 vs 68.5; Judge: 67.7 vs 65.2; Hard: 65.6 vs 75.7. Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.−1000100Easy: Jev 1.13.0: 93.9; Torchcast Decision 4B: 86.2Easy93.9 · 86.2Standard: Jev 1.13.0: 63.2; Torchcast Decision 4B: 68.5Standard63.2 · 68.5Judge: Jev 1.13.0: 67.7; Torchcast Decision 4B: 65.2Judge67.7 · 65.2Hard: Jev 1.13.0: 65.6; Torchcast Decision 4B: 75.7Hard65.6 · 75.7

Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.

Per-tier competence, the three request types pooled by their published decision counts.

Competence per tier — sealed set

Radar: competence per tier — sealed set, two systemsCompetence per tier — sealed set, Jev 1.13.0 vs Torchcast Decision 4B. Easy: 78.4 vs 77.4; Standard: 68.7 vs 70.8; Judge: 66.9 vs 65.3; Hard: 58.9 vs 62.1. Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.−1000100Easy: Jev 1.13.0: 78.4; Torchcast Decision 4B: 77.4Easy78.4 · 77.4Standard: Jev 1.13.0: 68.7; Torchcast Decision 4B: 70.8Standard68.7 · 70.8Judge: Jev 1.13.0: 66.9; Torchcast Decision 4B: 65.3Judge66.9 · 65.3Hard: Jev 1.13.0: 58.9; Torchcast Decision 4B: 62.1Hard58.9 · 62.1

Linear signed scale: −100 at the centre, 0 on the bold middle ring, 100 at the rim.

Per-tier competence on the sealed decisions, types pooled the same way; item text stays private.
How the categories were made

O1S chance-corrected competence: equal mean across the request types present, clipped to 0–100. Zero means the equal-weight mean across present request types was at or below baseline; negative means are clipped. It does not mean every type scored zero. Choice uses a random-option baseline; Noul uses a fixed 50% accuracy baseline (abstentions count as wrong); Score uses the midpoint-guess error (random-guess fallback when every gold is at the midpoint). Compare models within a category; row pools and supported types can differ.

Categories use ruled subject-topic labels and authoring use-case precedence across the answered pool union, with stable item identities counted once.

  • Official O1S; recorded failures count where supported; no missing response is imputed; stable identity dedupe parent first.
  • S-based category cells pool recorded supported responses from S+P and the answered L1/L2/L3 supplements. API overlay category cells keep their separately recorded A4/A5+P union. Headline and per-type/tier scores are unchanged.
  • Breakdown values are shown for cells with at least 30 scored items; more items are being added.
All values as a table
SpokeA: Jev 1.13.0B: Torchcast Decision 4B
The four score axes
Intelligence63.667.1
Calibration90.687.8
Speed91.292.4
Cost54.779.4
Capability by subject topic
Rules, policy & law49.473.0
Coding & software64.370.0
Math & numbers10.441.6
Finance & commerce40.978.0
Support & operations48.876.5
Everyday language77.545.2
Safety & security52.272.8
Use cases (TypeSafe categories)
Model routing73.076.2
Legal & compliance68.674.8
Customer support49.665.6
Other40.151.1
E-commerce40.566.6
Insurance claims46.1—
Risk assessment32.854.7
Financial crime35.836.1
Feature extraction28.3—
Lead generation49.266.9
Recruiting0.0—
Knowledge graphs28.8—
LLM guardrails68.0—
Moderation51.674.9
Code linting29.1—
Search & retrieval80.9—
Science21.2—
Advertising16.4—
Gaming5.1—
Demand forecasting0.0—
Competence per request type, open / sealed
Choice · open79.081.7
Choice · sealed75.873.0
Noul · open54.580.1
Noul · sealed46.170.9
Score · open63.347.5
Score · sealed63.049.8
Competence per tier — open set
Easy93.986.2
Standard63.268.5
Judge67.765.2
Hard65.675.7
Competence per tier — sealed set
Easy78.477.4
Standard68.770.8
Judge66.965.3
Hard58.962.1

Category radars count each answered item once from the pools named under each radar. API overlay rows use A4+P+L1+L2+L3; A5+P+L1+L2+L3; S+P+L1+L2+L3. Raw and unequated; cells under 30 answered items are omitted. Per-type and tier radars retain each row's original measurement pools: A4/A5 rows have 300 open plus 300 sealed items; full-set rows have S 1,200 plus P 300.

Languages

Raw chance-corrected competence per item language (0 = chance, 100 = perfect; can be negative), from each system's own measured items: Self-hosted rows use S 1,200 + P 300 + L1 (354 items) + L2 (33 items) + L3 (1,118 items), where answered. API rows re-run on A4/A5 use their 300/300-item sealed subset instead of S. Every row’s tag lists the pools it actually answered. L3 is a sealed language supplement (1,118 items), drawn 2026-10-07. Every language has at least 60 items in P ∪ L1 ∪ L2 ∪ L3. Header counts show the full S + P + L1 + L2 + L3 pool. Headline scores are unchanged; these raw cells are unequated, scored for language/category views only and outside the Composite. L3 items were written natively by Claude Sonnet 5.5, each solved blind and language-checked by GPT-6.1 Sol, with gold kept only when both agree or a second review confirms; no gold comes from Jev or any measured API. L3 is API-facing by design and is excluded from future headline draws. Rows with unfinished runs retain their actual coverage tags. L3 items were reviewed by an OpenAI model, so the L3-based language and category cells of OpenAI rows (OpenAI Decisions, GPT-6 Luna, GPT-5.6 Luna) carry that exposure; headline scores do not use L3. C1 adds English items for thin radar categories (everyday language, safety and the other use case). L4 adds 20 new Hindi items for Fastino GLiNER 2.5 Decide only. They were authored with OpenAI and independently solved and language-reviewed with Anthropic before sealing. Seventeen L4 requests succeeded; three HTTP 400 failures remain in the scored observations and do not count as completed coverage. L4 is outside headline and category scores and the common header counts. L5 adds 60 new items for OpenJev DeBERTa v3 Large only: 20 each in Arabic, Hindi and Greek. They were authored with OpenAI and independently solved and language-reviewed with Anthropic before sealing, then checked against the original 512-token native context before keeper selection. All 60 native requests succeeded. Previous failures remain scored and do not count as completed coverage. L5 is outside headline and category scores and the common header counts. Breakdown values are shown for cells with at least 30 scored items; more items are being added. Competence retains all supported scored observations, including failures and refusals. The mixed-language group is listed first, then English (1,306 items); the other 21 languages and the mixed-language group share 1699 items.

357 systems

JevBench v1.8.11 competence by item language and system
781306959088848383808077757575737372727171706965
Torchcast Decision 4B (v1.3)self-hosted · S+P (draw v1.6-regular-20261012)—6672—7168—————————————————
decide-1.1API · not measured847684747482788473818475797190758584—848684—
H2O-Lightning-4B v1.4.1self-hosted · not measured866875756859456481796969766475706860—868171—
Torchcast Decision 31B v1.0self-hosted · not measured909086899298909989919590979293877792—9494100—
Super DeciderAPI · not measured—8187—7686—————————————————
WeMM-JEVself-hosted · not measured908388848085728982838288948093896974—889690—
Myles-G 12BAPI · S+P (draw v1.6-regular-20261012)—5666—5986—————————————————
H2O-Lightning-4B v1.1self-hosted · S+P+L1+L2+L38563555960544938545841373446264946394855473123
Arcbright Elf 1 12B (r2)self-hosted · not measured776972697368398477757573836574786268—678363—
SD-JEV-4B-v12.1.0API · not measured906681726361416962837654786084657268—777873—
kettei-fatAPI · S+P (draw v1.6-regular-20261012)—8090—8687—————————————————
Wald 26B-A4Bself-hosted · not measured—6465—7472—————————————————
Jeff-1.1-Qwen-4Bself-hosted · not measured937067756587667284766366847478785047—878366—
Karpo-Decision-31Bself-hosted · not measured———————————————————————
z3-mini-soup-previewself-hosted · not measured896177715258346073706755634272695848—677156—
Decisio 12B tuned commit-thinkself-hosted · not measured836867676374577681717473786572716063—788063—
Torchcast Decision 4B (v1.2)self-hosted · S+P (draw v1.6-regular-20261012)—6065—7678—————————————————
iota-decide-4b-v0.4self-hosted · not measured846173726067455764665766825174755767—738558—
VisionJEV 1.0.0self-hosted · not measured868187838084678775907489897487895773—869288—
Fastino GLiDE (hosted API, 22 Oct 2026)API · S+P (draw v1.6-regular-20261012)—7674—7993—————————————————
d1-ultraAPI · S+P (draw v1.6-regular-20261012)—7066—6789—————————————————
Fastino GLiDE 1.2self-hosted · not measured—7774—7991—————————————————
Bobcat Flash 1.2self-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.8271454738543333514143373928264737485044582927
SubQ Snap 1.1API · S+P (draw v1.6-regular-20261012)—7575—8391—————————————————
H2O-Lightning-31B v1.0.7self-hosted · S+P (draw v1.6-regular-20261012)—8166—8391—————————————————
Microsoft-Decision-1 (updated weights, Oct 2026)API · not measured896978476480457686847556795880855271—826460—
StartLux-Decision v1.1 27B Instantself-hosted · S+P (draw v1.6-regular-20261012)—7278—7686—————————————————
Drex MaxAPI · S+P (draw v1.6-regular-20261012)—7972—8288—————————————————
baselabs mammoth-1self-hosted · not measured—7776—6590—————————————————
Decisio 12B tuned commitself-hosted · not measured—6070—6183—————————————————
Jeff-1.1-Mediumself-hosted · not measured—7582—8894—————————————————
Philly Decider v2self-hosted · not measured—5463—6670—————————————————
Perplexity Decider v1.1 27Bself-hosted · S+P+L1+L2+L38579505862634352705148425721386333676442643957
Bespoke Nimble-v4-maxAPI · S+P (draw v1.6-regular-20261012)—7170—7382—————————————————
Chon-1.0-E4Bself-hosted · not measured—5068—6883—————————————————
vLLM-SR d3-turboself-hosted · S+P (draw v1.6-regular-20261012)—7574—7683—————————————————
vLLM-SR d3self-hosted · S+P (draw v1.6-regular-20261012)—7473—7986—————————————————
Super Decider (thinking)API · not measured—8281—6788—————————————————
Decision-King v1.1self-hosted · not measured957483857793778486836486807491778081—818875—
KnowLine 26B-A4B-Gen2.5self-hosted · not measured—6775—6579—————————————————
Torchcast Decision 4B (v1.1)self-hosted · S+P (draw v1.6-regular-20261012)—5462—7261—————————————————
SubQ SnapAPI · S+P (draw v1.6-regular-20261012)—7468—7284—————————————————
KnowLine 35B-A3B-Gen1.5self-hosted · not measured—6171—5092—————————————————
basal-omni-1.0-ultraself-hosted · S+P (draw v1.6-regular-20261012)—7776—6586—————————————————
Xor 26B-A4Bself-hosted · S+P+L1+L2+L38061464048565136524434454332335040685050504246
torchcast-decision-27bself-hosted · S+P+L1+L2+L38174495851454340675242534723215038576139513544
Arcbright Elf 1 31Bself-hosted · not measured—7267—7486—————————————————
pplx-decider-v1.2self-hosted · S+P (draw v1.6-regular-20261012)—7378—7887—————————————————
Piccolo 4Bself-hosted · not measured856074705165425865706164715878675844—767168—
Quyet-1.0-Largeself-hosted · S+P+L1+L2+L38476475756665440665543525943296459676746554564
Nimble-v4-maxself-hosted · not measured—7469—7983—————————————————
Sage 1.3.0API · S+P+L1+L2+L38071485542494445575136444332264451656250593743
decider-12b v2self-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.8765515861404955614954435546435759675856405836
Xevn-4Bself-hosted · not measured—4963—4472—————————————————
Jeff-1.1-Largeself-hosted · not measured968885828086839898907881928091897889—968387—
kettei-chubAPI · S+P (draw v1.6-regular-20261012)—6277—6183—————————————————
Surogate Rune 26B-A4B v3 (v1.6 pool)self-hosted · S+P+L1+L2+L37559474336473737454433374538195237554340502333
Jeff-1.0-Largeself-hosted · S+P (draw v1.6-fastlane-20261009)—7189—6491—————————————————
Lychee-27Bself-hosted · not measured—7677—7888—————————————————
oenogarumself-hosted · not measured9277—69808165————7873—937676———9467—
decisio v0.8.0 on gemma-4-31B-itself-hosted · not measured———————————————————————
decider-12b v1, stock Gemma-4-12B-itself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.8562485049424654615142464738324944615049405143
vLLM-SR d3-miniself-hosted · not measured—5262—5673—————————————————
Sieve-31B-v2self-hosted · not measured—7167—6791—————————————————
MetaDecision-31B-Rainself-hosted · not measured98909496909276100939491781009296907888—10094100—
allec-oavgself-hosted · not measured9575—61786674————7370—887369———9482—
Piccolo 9Bself-hosted · not measured906475617479387167766575656176796656—807663—
d1API · S+P+L1+L2+L38066325534443336644028304128143327485141471232
Vera Largeself-hosted · not measured—6973—5880—————————————————
iota-decide-4b-v0.3self-hosted · not measured—4869—6470—————————————————
Umpire 35B-A3Bself-hosted · not measured785774625576327675625370615172624743—757561—
swanOneself-hosted · S+PThe pinned model weights cannot currently be accessed (Hugging Face returned HTTP 401). Supplement measurements remain pending access recovery; the radar shows the existing S + P items.—5786—48——————————————————
Piccolo 27Bself-hosted · not measured907984768885658876867983947790796285—879088—
Mercury DecideAPI · not measured———————————————————————
Winnow-12B Q8self-hosted · S+P+L1+L2+L37864405939523954534541294433414636575534384440
cribl-decision-1.0-27Bself-hosted · not measured906468646275637859695866874876675264—837472—
Bobcat 1.3self-hosted · not measured—6971—6387—————————————————
JPT-35B-A3Bself-hosted · S+P+L1+L2+L3865639483245342843473228441763934284633451720
Chon-1.0-12B-Fastself-hosted · not measured—5671—7888—————————————————
baselabs elephant-1self-hosted · not measured—5374—6272—————————————————
metask-jev-rain-31B-trainedself-hosted · S+P (draw v1.6-regular-20261012)—7169—6390—————————————————
deck-31Bself-hosted · S+P+L1+L2+L39178606955686560747156587150536860706159554560
Cygnetself-hosted · S+P+L1+L2+L37957385641413239634534274236214245535434394640
Minimodels Decision 1.0self-hosted · not measured948083778290718766868983867988846185—868173—
metask-jev-rain-12B-trainedself-hosted · S+P (draw v1.6-regular-20261012)—6474—7285—————————————————
DOPEST 1.0 31Bself-hosted · not measured—7460—6991—————————————————
Eikos-27Bself-hosted · S+P+L1+L2+L37568455440465644505838414626304139515343504249
Jev 1.13.0API · S+P+L1+L2+L3766838442248353436363015431173420364323471921
Jeffan-31b-1710self-hosted · not measured—7671—6777—————————————————
Kev 27Bself-hosted · S+P+L1+L2+L37769475038424244644441454420265031515535482233
Gemini Decision 1API · not measured937382676972787878606268916886646674—857579—
Jev-Omniself-hosted · S+P+L1+L2+L38656385859444537585038274928255038575547484637
Celeris-1 DecisionAPI · S+P (draw v1.6-regular-20261012)—5959—5688—————————————————
GEV-26B-Decideself-hosted · S+P+L1+L2+L37852333731423742372831233028213529422140282725
decisio v0.8.0 on gemma-4-12B-itself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.7956395539423338634129253435253744485638404227
Myles-G 31BAPI · S+P (draw v1.6-regular-20261012)—7571—6889—————————————————
RSI-Jev v6.1-VL 27Bself-hosted · S+P (draw v1.6-regular-20261010-a2)—57——5172———32———————53—————
Nagi-ENORMOUS 27Bself-hosted · S+P (draw v1.6-regular-20261012)—6266—7784—————————————————
StartLux-Decision v1.1 27B AutoThinkself-hosted · S+P (draw v1.6-regular-20261012)—6671—6184—————————————————
Jev 1.14.0API · S+P (draw v1.6-regular-20261012)—6373—4275—————————————————
metask-jev-rain-31B-trained service v3self-hosted · S+P (draw v1.6-regular-20261012)—7468—6382—————————————————
decisio v0.9.0 on gemma-4-12B-itself-hosted · S+P+L1+L2+L38156415538403141624128293634253744485637404331
Bespoke Nimble 9B v3self-hosted · S+P+L1+L2+L3806145503544303056412774032173623324260542037
AutoTrust-Decider-Pro-27Bself-hosted · S+P+L1+L2+L3716236422341363754393337439194737454737432648
torchcast-decision-12bself-hosted · S+P+L1+L2+L38665496447483953645044474544335440685354445844
Messier One v0.4self-hosted · S+P (draw v1.6-regular-20261012)—5168—7072—————————————————
shisa-de-1self-hosted · S+P (draw v1.6-regular-20261012)—4561—4588—————————————————
Microsoft-Decision-1 (Azure Foundry)API · not measured———————————————————————
Clefself-hosted · S+P+L1+L2+L38265504927493846554533434321303136524437492336
wity-1 (Wity, reasoning always)API · S+P+L1+L2+L3757357683250334446432940173632294144554331185
wity-1 (Wity, reasoning auto)API · S+P+L1+L2+L38373596331602940333844333954241642563936451516
Wald 4B v2.1self-hosted · S+P (draw v1.6-regular-20261010-a2)—50——3455———49———————25—————
Hopper 12B trainedself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.7954505142483339544232233828163632484734322434
Blink v0.3 26B-A4Bself-hosted · S+P (draw v1.6-regular-20261010-a2)—46——4570———25———————39—————
Decision 2.0 Vega 27Bself-hosted · S+P+L1+L2+L36864394923362630353641304418222930465121432140
René-1 31B FP8self-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.7567453942464031553735375413254433474342462043
AutoJev-27B (denis-pplx, Qwen3.8-27B)self-hosted · S+P+L1+L2+L37459413727353332513830244216182736424832452527
OpenJev (thinking, BF16)self-hosted · S+P+L1+L2+L38778595431553160646653506543536046536251564142
Jebadiah 27Bself-hosted · S+P+L1+L2+L36564364730342239514129183718113133494629441726
decider chat on Gemma-4-31B-itself-hosted · S+P+L1+L2+L37564585434535045565945395245324344514443444449
attest-31B-v1.0API · S+P (draw v1.6-regular-20261012)—6767—5889—————————————————
NInfer Qwen3.8-Flash-Next mixedself-hosted · S+P+L1+L2+L37253415024333033553328164320273635543138432520
Metask rain 4Bself-hosted · S+P (draw v1.6-fastlane-20261009)—4865—6062—————————————————
Sage 1.4API · S+P (draw v1.6-regular-20261012)—6565—6583—————————————————
FrontiersMind Lumma-fev-27Bself-hosted · not measured—5564—5176—————————————————
Respan Span-02self-hosted · S+P (draw v1.6-regular-20261012)—6050—6180—————————————————
NInfer Qwen3.8-27B NVFP4self-hosted · S+P+L1+L2+L3725545513540322341423322382217183758372643178
SPX-CD FlashAPI · S+P+L1+L2+L3786338433640273449483122301783824375146501229
decider-35b-a3bself-hosted · S+P+L1+L2+L377523129164523233724291644952422413624381428
Diffusion Jevself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.8259515749653741524448294345364944635050513816
SPX-CD-Omniself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.8053354046362932504441344426314822445635382434
OpenAI DecisionsAPI · A5+P+L1+L2+L3L3 items were reviewed by an OpenAI model, so the L3-based language and category cells of this row carry that exposure; headline scores do not use L3.7659364841552737433724234637303327403339374025
InstinctAPI · A4+P+L1+L2+L3675822353228273152482018341923223939394243137
HS1 0.1.2API · S+P (draw v1.6-regular-20261012)—5054—4678—————————————————
KnowLine 31B-Gen1self-hosted · not measured—8172—7588—————————————————
Amazon Strands Decider 26B-A4Bself-hosted · not measured—4368—4969—————————————————
Turion Jev 31Bself-hosted · not measured—7572—6283—————————————————
KnowLine 4B-Gen5.5self-hosted · not measured—4964—2776—————————————————
JevOneself-hosted · S+P+L1+L2+L37349373522331021323122102010920243040183210
Seb-9B-v2self-hosted · not measured—4465—3168—————————————————
pplx-decider-flash-v1.2self-hosted · S+P (draw v1.6-regular-20261012)—4766—5474—————————————————
APUS-OpenJev-v1-9Bself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.7448333125371822374180202561628293034351712
Aplomb 1self-hosted · S+P+L1+L2+L3The headline keeps the original published measurement. Its originally pinned Hugging Face revision (f8e2dc81) was later deleted upstream, so the language and category supplements (L1, L2, L3) were measured on 12 Oct 2026 with the current upstream revision d6743291 (all file hashes recorded) on an isolated pod. That revision reproduces the original answers: 98 % agreement on the 300 public items (published bar 95 %) and 97.5 % on all 1,500 headline items. Each supplement carries the original P300 answers by stable item identity; they count once. All new (sealed) supplement items were answered; the only failed requests (3 public items, GPU out-of-memory) are covered by the original P300 answers. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.694643362641212244272214242811272725433338132
JADEself-hosted · S+P+L1+L2+L320000000000000000000000
Tern-35B-A3Bself-hosted · not measured—4367—3676—————————————————
AJev lm6qself-hosted · not measured—7468—6784—————————————————
NVIDIA Nemotron 3.5 Super VLAPI · not measured856090835473657660584478773973413573—716250—
Arcbright Elf 1 12Bself-hosted · S+P (draw v1.6-regular-20261012)—4153—5779—————————————————
Plumb-4Bself-hosted · S+P+L1+L2+L3724535453630382732342583240142431252827392220
wity-1 (Wity, reasoning off)API · S+P+L1+L2+L367523848293621261013171417176112430313333261
SPX-CD ProAPI · S+P+L1+L2+L38270445537445343625035474119224737485429522947
Messier One v0.2self-hosted · not measured———————————————————————
spark-s1-4b-v6self-hosted · S+P+L1+L2+L380463331182891737352202223015292023102627
Bev / Bonsai 27Bself-hosted · S+P+L1+L2+L367462635193014172522160900024286212746
Clef-Flashself-hosted · S+P+L1+L2+L3694440318349194129231810201112232323361311
Quyet-1.0-Mediumself-hosted · S+P+L1+L2+L360443332254225153532238152002933351921351214
Open-Jev 9Bself-hosted · S+P+L1+L2+L370474048234426212635180245142819351530381417
LitJevself-hosted · S+P+L1+L2+L369493439273630233434280242210103239402930131
Bespoke Nimble 9Bself-hosted · S+P+L1+L2+L36645413635411021372619423912625263940321411
SimpleJev Qwen3.8-27B (self-hosted, v1.6 pool)self-hosted · S+P+L1+L2+L38062445234414329614133243925293241495031502926
Qwen3.5-9B Jev-like data-mix v2self-hosted · S+P+L1+L2+L37146292418376133531120253091830382424321
cribl-decision-1.0-9Bself-hosted · not measured774665404268344758483144454865623942—604940—
levself-hosted · S+P+L1+L2+L35541282022221222282229822112141728192437537
Strike-1API · not measured674670562858286657563443684262545257—735442—
JPT-4Bself-hosted · S+P (draw v1.6-regular-20261012)—3675—2775—————————————————
jev-localself-hosted · S+P+L1+L2+L37146433121312122333217720127212910342529115
Aleria Judge 1.1.1self-hosted · not measured—7067—6585—————————————————
Saif 4Bself-hosted · not measured664458512756265866343046514749543621—676442—
Bespoke Nimble-v4API · S+P (draw v1.6-regular-20261012)—4267—3973—————————————————
decider-4b v2self-hosted · S+P+L1+L2+L36242412511288222132320181318183222242911
Jeffan-4B-2010self-hosted · not measured—3966—3063—————————————————
ClassOne Qwen 3.5 9Bself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.714538512636231930432301781411301718401331
Vansa-3.4API · A4+P+L1+L2+L35447124228171210244311013191291726153335011
Wald 4Bself-hosted · S+P (draw v1.6-fastlane-20261009)—4351—3554—————————————————
Granite Switch 4.1 30Bself-hosted · not measured845163594161295459533464582663632262—598159—
Open-Jev 27B v1.1self-hosted · S+P+L1+L2+L3645544412240242144332837388112311423634372423
GPT-6 Luna (low reasoning effort)API · A4+P+L1+L2+L3L3 items were reviewed by an OpenAI model, so the L3-based language and category cells of this row carry that exposure; headline scores do not use L3.9796919391941009410099989694939694969710098969997
GLiClass v4 9Bself-hosted · not measured—3658—5980—————————————————
Autoloops – Gemma 4 31B ITAPI · A4+P+L1+L2+L37367445044566142616043484848476050554945534151
metask-jev-4bself-hosted · S+P+L1+L2+L37340303415211619282412915409112316122640
cribl-decision-2self-hosted · not measured714467374056164264412749545057634034—656435—
Kahn1 4Bself-hosted · S+P+L1+L2+L3674036462835253029452553525211433192841302020
JevK5 v0.3self-hosted · S+P+L1+L2+L36743404229192013283018028278162817223429318
GPT-6 Luna (default medium reasoning effort)API · A4+P+L1+L2+L3L3 items were reviewed by an OpenAI model, so the L3-based language and category cells of this row carry that exposure; headline scores do not use L3.9797989691981009999969810096959895979995961009996
deck-4B v1.0self-hosted · S+P+L1+L2+L367433745322221192930220312651323182736291917
Seb-9Bself-hosted · not measured———————————————————————
Decision-4Bself-hosted · S+P+L1+L2+L36143271903107202226261000211713152500
decisor-4bself-hosted · S+P (draw v1.6-fastlane-20261009)—3670—4458—————————————————
janus 4Bself-hosted · S+P+L1+L2+L36340352315199152120162259001129171520107
Jebadiah 9B v2.1self-hosted · S+P (draw v1.6-regular-20261012)—3567—3270—————————————————
JevK5 v0.2.0self-hosted · S+P+L1+L2+L365362636212123152420601013742421191826132
Amos 1 mediumAPI · S+P (draw v1.6-regular-20261012)—3562—4686—————————————————
Decision 2.0 Lux 9Bself-hosted · S+P (draw v1.6-regular-20261012)—3164—4067—————————————————
Verdin-4Bself-hosted · S+P (draw v1.6-regular-20261012)—3156—4054—————————————————
TypeCastLM 1.4.0self-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.57382425822161728181201790517961519133
blink-4bself-hosted · S+P (draw v1.6-regular-20261012)—3173—3179—————————————————
Fastino GLiDEAPI · A4+P+L1+L2+L36973405329424337715243334322333756534745433540
Imajev-9Bself-hosted · S+P (draw v1.6-regular-20261012)—3363—4065—————————————————
jiwo 4bself-hosted · S+P (draw v1.6-regular-20261010-a2)—30——3443———12———————12—————
Imajev-4Bself-hosted · S+P+L1+L2+L351373027924019121014090008101352910
Uniphore-4B-Deciderself-hosted · S+P (draw v1.6-regular-20261012)—3271—2570—————————————————
Malkuth-4Bself-hosted · S+P+L1+L2+L3493721173117177540700000422000
Hopperself-hosted · S+P+L1+L2+L35634312215220151612130141230121517151100
SimpleJev Qwen3.6-35B-A3BAPI · A4+P+L1+L2+L36563374226433041535424344417222134302034422520
Decisio v0.9.0 on Qwen3.6-35B-A3Bself-hosted · not measured—3157—3373—————————————————
Clef-omniself-hosted · S+P (draw v1.6-regular-20261010-a2)—35——3652———18———————45—————
JEB-35B-A3Bself-hosted · S+P (draw v1.6-regular-20261012)—2949—4064—————————————————
jqvself-hosted · S+P+L1+L2+L3703916241829817151721010501872716252903
CoCo-Decision-4Bself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.5136262321171822281716014902724162025121
Decision 4B v1.2self-hosted · S+P+L1+L2+L3653636321625162227342102432222728203534251214
Manchego v2.1self-hosted · S+P+L1+L2+L3503325110150191281702080212196473
Standard One 3B SH v4self-hosted · not measured—3247—2769—————————————————
Sors-2Bself-hosted · not measured64404223394331344838155469132443300—584817—
Jev-Style-2Bself-hosted · not measured5733362629382142281492916211127100—425727—
Raw Qwen3 4B Instruct 2507 direct logitsself-hosted · S+P+L1+L2+L359341731532017244170103192142519213104
Standard One 3B SHself-hosted · S+P (draw v1.6-regular-20261012)—2872—2077—————————————————
GPT-5.6 LunaAPI · A4+P+L1+L2+L3L3 items were reviewed by an OpenAI model, so the L3-based language and category cells of this row carry that exposure; headline scores do not use L3.9692869284869290869490799397948987848591878290
Jebadiah 9B v2self-hosted · S+P (draw v1.6-regular-20261012)—2767—3167—————————————————
reflex 4Bself-hosted · S+P+L1+L2+L356333324413113151211061117017172941804
typecastlmself-hosted · S+P+L1+L2+L3542930196246112423130571081514820133
Amos 1 largeAPI · S+P (draw v1.6-regular-20261012)—7071—7784—————————————————
iota-decide-4bself-hosted · S+P (draw v1.6-regular-20261012)—2470—1668—————————————————
Decision 4B v1.1self-hosted · S+P+L1+L2+L3593329391224102124411701519111822727282255
Standard One 8B SHself-hosted · S+P (draw v1.6-regular-20261010-a2)—30——4145———9———————4—————
kev 8Bself-hosted · S+P+L1+L2+L351352426922087070040002092700
Compass 0.2.0self-hosted · S+P (draw v1.6-regular-20261012)—1045—1634—————————————————
Granite Switch 4.1 8Bself-hosted · not measured62344637325422404723227322933492038—406627—
pjev-2bself-hosted · S+P (draw v1.6-regular-20261012)—2553—3552—————————————————
Instinct Dual 4BAPI · A4+P+L1+L2+L3363212500027132000001161201500
ArseneLupin v1.1self-hosted · S+P (draw v1.6-regular-20261012)—1661—2040—————————————————
Gemini 3.1 Flash-LiteAPI · A4+P+L1+L2+L38468426457535846635842455551274340493554544944
Standard One 8Bself-hosted · S+P+L1+L2+L362382217201712426197012304023142334610
Arbiter-4Bself-hosted · S+P (draw v1.6-regular-20261012)—2759—2066—————————————————
system-one-openAPI · A4+P+L1+L2+L3233181491113167166000001910022502
Decider-4Bself-hosted · S+P (draw v1.6-regular-20261012)—2162—1354—————————————————
SemIf, formerly OpenJevself-hosted · S+P+L1+L2+L3583015162225101415901900011157112000
RYOTIDE-Qwen9self-hosted · S+P (draw v1.6-fastlane-20261009)—1970—2960—————————————————
ClassOne Gemma 4 E4Bself-hosted · S+P (draw v1.6-regular-20261010-a2)—22——5021———0———————34—————
JEV Qwen3.5-9B Base NVFP4self-hosted · S+P+L1+L2+L34027241261611211700700025130705
DecisionNode Edge 0.9.2self-hosted · not measured—2047—4869—————————————————
openjev-sglangAPI · A4+P+L1+L2+L36047202919280192442191212139133224143728200
ClassOne Qwen 3.5 4Bself-hosted · S+P (draw v1.6-regular-20261010-a2)—21——3123———1———————9—————
system-oneself-hosted · S+P+L1+L2+L3513162743671719600700101618011300
Jez-large-1.0.0self-hosted · not measured—948—2362—————————————————
Chon-1.0-12Bself-hosted · S+P (draw v1.6-regular-20261012)—634—6250—————————————————
RYOTIDE-Qwenself-hosted · S+P (draw v1.6-regular-20261012)—2275—3373—————————————————
Nomaself-hosted · S+P (draw v1.6-regular-20261012)—2057—1931—————————————————
Amber 4Bself-hosted · not measured5728—1730290————163—365129———4740—
Raw Qwen3 8B direct logitsself-hosted · S+P+L1+L2+L3543071503001117718816000122751400
local-jev Qwen3.5-4Bself-hosted · S+P+L1+L2+L33515111105028030000078001100
baselabs joey-1self-hosted · not measured4428372963003145081801613800—35342—
OpenSourceJevself-hosted · S+P+L1+L2+L33320185211050000000000001000
Decision 2Bself-hosted · S+P+L1+L2+L320147001000000000003000000
open-alternative-jevself-hosted · S+P+L1+L2+L34324203014373911003007274401
Krino-2Bself-hosted · S+P (draw v1.6-regular-20261012)—1612—4345—————————————————
Liquid AI d1-3Bself-hosted · S+P (draw v1.6-regular-20261010-a2)—14——306———0———————1—————
Open-Jev 2Bself-hosted · S+P+L1+L2+L339181116171009712000000003600
classify-1API · not measured—08—69—————————————————
Imajev-2Bself-hosted · S+P (draw v1.6-regular-20261012)—547—1742—————————————————
decider-2bself-hosted · S+P+L1+L2+L34225770142432802300581200700
D1A on Gemma 4 E4Bself-hosted · not measured—841—3143—————————————————
Nemotron Diffusion 8Bself-hosted · S+P+L1+L2+L330160016190060004012001800300
Nev-2Bself-hosted · not measured———————————————————————
DeepSeek V4.1 FlashAPI · A4+P+L1+L2+L39385847137836597687279658680785979859082876257
kev 4Bself-hosted · S+P+L1+L2+L344252811014051907016000013001600
cribl-decision-2-miniself-hosted · not measured243272111801110080005100—18310—
DecisionNode Edge Mini 0.9.2self-hosted · not measured—338—4546—————————————————
Malkuth-2Bself-hosted · S+P+L1+L2+L33514202600000000000000400
Qwen3-Reranker-4Bself-hosted · S+P+L1+L2+L340000300000000000000000
jiwo 0.8bself-hosted · S+P (draw v1.6-regular-20261010-a2)—0——257———0———————0—————
ZeroEntropy zerank-2self-hosted · S+P+L1+L2+L300000000000000000000000
Vega 4Bself-hosted · S+P (draw v1.6-regular-20261010-a2)—8——434———6———————0—————
EXAONE-4.0-1.2B-JEV v0.3self-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.56000000000000000000000
SimpleJev Qwen3.8-27BAPI · A4+P+L1+L2+L37063284038433322544323253417312144433731552927
Raw Phi-4 mini direct logitsself-hosted · S+P+L1+L2+L313265000010007000010001502
decision-machine-1API · A4+P+L1+L2+L3104000001000050000004005
SimpleJevself-hosted · S+P+L1+L2+L3681100000000000000000960
Jeetu Decider 2Bself-hosted · not measured—06—2023—————————————————
JevActAPI · A4+P+L1+L2+L312001000000000000001001400
D1A on Gemma 4 E2Bself-hosted · not measured—05—326—————————————————
smalljev semantic-v9self-hosted · S+P+L1+L2+L300000000000000000000000
Gutsy 0.8B v0.3self-hosted · S+P (draw v1.6-regular-20261010-a2)—0——122———0———————0—————
Vega-U (Qwen3.5-4B reader)self-hosted · S+P (draw v1.6-regular-20261012)—00—1010—————————————————
PurpleMIST-Mini-1.0self-hosted · S+P (draw v1.6-regular-20261012)—00—00—————————————————
kev 0.6Bself-hosted · S+P+L1+L2+L310000000000000000000000
jul fastself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.00000000000000000000000
OpenDecisionself-hosted · S+P+L1+L2+L301000000000000000000000
ClassOne Gemma 4 E2Bself-hosted · not measuredThis model's server gives different answers after each restart, so a supplement run cannot reproduce the row's original answers (87.7 % agreement on the 300 public items, below the 95 % bar). Cause (verified 12 Oct 2026): the published model has no trained embeddings for its 12 decision delimiter tokens, so they are randomly initialised at every server start. Its language and category supplements are not published; the radar shows the existing S + P items.———————————————————————
Qwen3.5-0.8B Decision Modelself-hosted · not measured———————————————————————
Fastino GLiNER-2.5-DecideAPI · S+P+L1+L2+L3———————————————————————
Deem 0.8B v1self-hosted · S+P+L1+L2+L350000100000000000000800
Decision Fastself-hosted · S+P+L1+L2+L300000000000000000000000
Quyet-1.0-Small-ENself-hosted · S+P+L1+L2+L300000000000000000000000
ClassOne Qwen 3.5 2Bself-hosted · S+P (draw v1.6-regular-20261010-a2)—3——180———0———————11—————
Laya typed-decisionsself-hosted · S+P+L1+L2+L300000000000000000000000
Vega-U (Qwen3.5-0.8B reader)self-hosted · S+P (draw v1.6-regular-20261012)—00—00—————————————————
WaterSheepself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.00000000000000000000000
openJev Verdict 1.4self-hosted · S+P+L300000000000000000000000
Raw Qwen3 0.6B direct logitsself-hosted · S+P+L1+L2+L30000000010000000130400800
lev-350mself-hosted · S+P+L1+L2+L300000000000000000000000
Raw Qwen3 1.7B direct logitsself-hosted · S+P+L1+L2+L3100060000200000000000000
kev 0.5Bself-hosted · S+P+L1+L2+L300000000000000000000000
Jez-small-1.0.0self-hosted · not measured—00—00—————————————————
openJev Verdictself-hosted · S+P+L1+L2+L300000000000000000000000
Quyet-1.0-Smallself-hosted · S+P+L1+L2+L300000000000000000000000
Reflex Instinct 0.6Bself-hosted · not measured———————————————————————
watt-flash-0.1self-hosted · S+P+L1+L2+L300000000000000000000000
Quyet-1.0-Tinyself-hosted · S+P+L1+L2+L300000000000000000000100
Phocinae-Largha-150M-v1self-hosted · S+P (draw v1.6-regular-20261012)—00—00—————————————————
Liquid AI d1-omni-600Mself-hosted · S+P (draw v1.6-regular-20261010-a2)—0——00———0———————0—————
Vega 0.8Bself-hosted · S+P (draw v1.6-regular-20261010-a2)—0——00———0———————0—————
open-jev-deberta-v3-largeself-hosted · S+P+L1+L2+L3———————————————————————
modernbert-jev v2self-hosted · not measured———————————————————————
Julia-1self-hosted · S+P (draw v1.6-regular-20261012)—00—022—————————————————
verdict-smallself-hosted · S+P+L1+L2+L300000000000000000000001
Tacet Sonataself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.20000000000000000000000
Layaself-hosted · S+P+L1+L2+L300000000000000000000000
Mixedbread mxbai-rerank-base-v2self-hosted · S+P+L1+L2+L300000000000000000000000
Laya multilingualself-hosted · S+P+L1+L2+L3000000000000005141230000
Certo v1self-hosted · S+P+L1+L2+L300000000000000000000000
CLM-8Bself-hosted · S+P+L1+L2+L300000000000000000000000
BAAI bge-reranker-v2-m3self-hosted · S+P+L1+L2+L300000000000000000000000
Alibaba GTE Reranker ModernBERT-baseself-hosted · S+P+L1+L2+L300000000000000000000000
Mirrorself-hosted · S+P+L1+L2+L300000000000000000000000
Open Jev JSON Canvasself-hosted · S+P+L1+L2+L37665554247563755605551403446443341504754514953
Qwen3.8 27BAPI · A4+P+L1+L2+L31124000000000000000000000
APUS-OpenJev-v1-35B-A3Bself-hosted · S+P+L1+L2+L3Language and category cells use the unchanged original S+P observations and the newly completed frozen supplemental items (L1, L2, L3), answered with the original model and serving setup on an isolated pod. Each supplement carries the same original P300 answers by stable item identity; they count once in the union. Failed requests stay scored as failures and are not imputed. Headline, rank, Composite and original pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.8254374047402725414734142333123132373935371710
BB-Qwen3.5-4B-LoRAAPI · not measured———————————————————————
Bongard-miniself-hosted · S+P (draw v1.6-regular-20261012)—434—2643—————————————————
DecisionNode FlashAPI · not measured764769594155386253454143544860524556—656646—
DecisionNodeAPI · not measured826179617076417868685170655977616270—747864—
jevora-9b-v1self-hosted · S+P (draw v1.6-regular-20261012)—1439—2358—————————————————
Kev-9B v2self-hosted · S+P (draw v1.6-regular-20261012)—3044—3375—————————————————
Nagi-HUGE 12Bself-hosted · S+P (draw v1.6-regular-20261012)—5373—5784—————————————————
Nemotron Deciffusion 8B v2self-hosted · S+P (draw v1.6-regular-20261012)—5068—5460—————————————————
PurpleMIST-Flash-1.0self-hosted · S+P (draw v1.6-regular-20261012)—3557—4877—————————————————
RYOTIDE-Gemmaself-hosted · S+P (draw v1.6-regular-20261012)—3043—4765—————————————————
TDE general large v0.1self-hosted · S+P (draw v1.6-regular-20261012)—00—00—————————————————
Trio-Spark v1.3API · S+P (draw v1.6-regular-20261012)—2945—1357—————————————————
AutoJev-27B (RTX PRO 6000)self-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.7460423827353332514030224015182837464832452527
Bosun v3.1 0.6Bself-hosted · S81+P299+L1+L2+L3Catalogue entry · measured language cells40000000000000000000700
GLiNER2self-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. About 1,000 issued items were lost to out-of-memory kills and a machine shutdown; they stay counted as spent and are never re-asked, so 17 of 23 languages rest on 50 to 59 answered items instead of the 60-item target. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.110000000140000000000120010
GLiNER2 largeself-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.791000000000970001008800
GLiNER2.5 multiself-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.00000000000000000000400
GLiNER2.5 smallself-hosted · P+L1+L2+L3Catalogue entry · measured language cells00000000000000000000000
jeffself-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.00000000000000000000000
Jobe Qwen3.5-4Bself-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.58301416125491415801900011156132000
mica-v01-4bself-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.5538322232602218264125441392230122100
Needle 3self-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from new runs of the supplements L1, L2 and L3 (all items) on the original pinned closed engine, approved by the benchmark owner, on an isolated machine, plus 281 S+P items: 153 from an earlier partial run and 128 never-asked items chosen by their use-case label ('other') to complete that radar spoke (these items also count in the other cells, so the S+P part is not a random sample); the rest of S+P was not run. The engine abstains on about 30 % of items and returns invalid UTF-8 on about 7 % (mostly non-Latin scripts); both count as unanswered. Its chance-corrected competence is below zero in every cell, which the table shows as 0. Its answers to the same public items differ between machines for about 10 % of items. 22 of 23 languages rest on 19 to 59 answered items instead of the 60-item target; the radars are complete. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.0000000000—0000—0—00—0—
Needle 3, options as toolsself-hosted · S+P+L1+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells combine an earlier partial run on the current pools (175 S+P items) with a new run of the supplements L3 (1,382 of 1,418 items) and L1 (329 of 654) on the original pinned closed engine, approved by the benchmark owner, on an isolated machine; L2 and the rest of S+P were not run before the machine's time limit. The engine abstains on about 9 % of items and returns invalid UTF-8 on about 7 % (mostly non-Latin scripts); both count as unanswered. Its chance-corrected competence is below zero in every cell, which the table shows as 0. Its answers to the same public items differ between machines for about 10 % of items. 18 of 23 languages rest on 24 to 59 answered items instead of the 60-item target; the radars are complete. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.004000000000000—000003—
NInfer Qwen3.8-27B NVFP4 (T=1.5)self-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.715039512432292038372715371918183351262538145
OpenJev (DiffusionGemma 26B-A4B NVFP4, razorback16)self-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. This model samples stochastically: on the 300 public items its answers in the supplement passes agree with its S+P pass on 86-89 %; those public items count once, from the S+P pass. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.6054464149462937424437303023273128332735313334
reflex-27bself-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.685634382736322047492920282311183137374030135
Surogate Rune 26B-A4B v3 (RTX PRO 6000)self-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells combine its S+P run on the current pools (6 Oct 2026) with new L1, L2 and L3 runs on the same pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.7559474336473737454433374538195237554340502333
Vonself-hosted · S+P+L1+L2+L3Catalogue entry · measured language cellsThis historical row keeps its published v1.5 headline. Its language and category cells come from a new measurement on the current item pools (S+P, L1, L2, L3), run with the original pinned model and serving setup on an isolated machine; each item was sent once and failed requests stay scored as failures. Headline, rank, Composite and pricing are unchanged. L3 items were reviewed by an OpenAI model, so cells based on L3 carry that exposure; headline scores do not use L3.00000000000000000000000
Wrappers (listed, never ranked)
classifier.devAPI · A4+P+L1+L2+L3737223353643342833412315359183225444417511434
metask-jev-rain-12Bself-hosted · S+P (draw v1.6-fastlane-20261009)—6479—5583—————————————————

Intelligence gate and Noul decisiveness

The Composite applies a soft gate below Intelligence 50 (× (I/50)²). Rows under it show their value with a "below gate" label. The decisive rate is the share of valid Noul answers with P(yes) ≥ 0.80 or ≤ 0.20 (for systems that return only a label, every valid label counts as decisive); only decisive answers count as answers. Accuracy among decisive answers is a diagnostic and does not enter any score.

SystemIntelligenceChoiceNoulScoreNoul decisive rateAccuracy among decisive
Torchcast Decision 4B67.1nativenativenative100 %86 %
decide-1.172.8nativenativenative100 %95 %
Torchcast Decision 31B v1.085.5nativenativenative100 %97 %
Super Decider78.9nativenativenative100 %93 %
Myles-G 12B57.3nativenativenative100 %79 %
H2O-Lightning-4B v1.160.0nativenativenative100 %85 %
Arcbright Elf 1 12B63.9nativenativenative100 %89 %
SD-JEV-4B-v12.1.063.3nativenativenative85 %97 %
kettei-fat79.8nativenativenative95 %96 %
Wald 26B-A4B65.5nativenativenative100 %82 %
Jeff-1.1-Qwen-4B65.3nativenativenative100 %91 %
Karpo-Decision-31B80.2nativenativenativenot supported—
z3-mini-soup-preview57.5nativenativenative78 %96 %
Decisio 12B tuned commit-think65.7nativenativenative100 %89 %
Torchcast Decision 4B60.3nativenativenative100 %84 %
iota-decide-4b-v0.457.3nativenativenative100 %86 %
Fastino GLiDE76.1nativenativenative86 %97 %
d1-ultra69.7nativenativenative86 %94 %
Fastino GLiDE 1.276.4nativenativenative86 %98 %
Bobcat Flash 1.265.6nativenativenative89 %92 %
SubQ Snap 1.175.8nativenativenative89 %95 %
H2O-Lightning-31B v1.0.779.3nativenativenative100 %92 %
Microsoft-Decision-164.2nativenativenative83 %95 %
StartLux-Decision v1.1 27B Instant73.2nativenativenative82 %99 %
Drex Max78.9nativenativenative93 %94 %
baselabs mammoth-176.2nativenativenative89 %97 %
Decisio 12B tuned commit62.8nativenativenative100 %87 %
Jeff-1.1-Medium74.5nativenativenative100 %89 %
Philly Decider v256.1nativenativenative100 %80 %
Perplexity Decider v1.1 27B75.6nativenativenative91 %95 %
Bespoke Nimble-v4-max68.2nativenativenative78 %98 %
Chon-1.0-E4B55.5nativenativenative100 %80 %
vLLM-SR d3-turbo74.8nativenativenative87 %96 %
vLLM-SR d372.9nativenativenative86 %96 %
Super Decider80.0nativenativenative100 %93 %
Decision-King v1.172.8nativenativenative94 %95 %
KnowLine 26B-A4B-Gen2.565.2nativenativenativeunavailableunavailable
Torchcast Decision 4B54.2nativenativenative100 %78 %
SubQ Snap72.3nativenativenative86 %95 %
KnowLine 35B-A3B-Gen1.560.3nativenativenative75 %95 %
basal-omni-1.0-ultra74.1nativenativenative92 %94 %
Xor 26B-A4B57.5nativenativenative81 %94 %
torchcast-decision-27b70.4nativenativenative83 %97 %
Arcbright Elf 1 31B71.9nativenativenative100 %91 %
pplx-decider-v1.271.7nativenativenative84 %96 %
Piccolo 4B56.3nativenativenative82 %94 %
Quyet-1.0-Large73.4nativenativenative91 %93 %
Nimble-v4-max71.0nativenativenative84 %97 %
Sage 1.3.065.6nativenativenative83 %95 %
decider-12b v261.9nativenativenative99 %83 %
Xevn-4B52.1nativenativenative100 %79 %
kettei-chub62.2nativenativenative87 %94 %
Surogate Rune 26B-A4B v356.1nativenativenative73 %97 %
Jeff-1.0-Large71.9nativenativenativenot supported—
Lychee-27B74.3nativenativenative92 %94 %
oenogarum71.8nativenativenot supported86 %99 %
decisio v0.8.0 on gemma-4-31B-it70.4nativenativenative94 %96 %
decider-12b v1, stock Gemma-4-12B-it59.6nativenativenative96 %87 %
vLLM-SR d3-mini50.8nativenativenative77 %91 %
Sieve-31B-v268.7nativenativenative82 %98 %
MetaDecision-31B-Rain85.0nativenativenative100 %97 %
allec-oavg70.5nativenativenot supported86 %99 %
Piccolo 9B60.9nativenativenative87 %94 %
d162.1nativenativenative83 %93 %
Vera Large67.8nativenativenative82 %95 %
iota-decide-4b-v0.352.5nativenativenative100 %79 %
Umpire 35B-A3B55.3nativenativenative83 %95 %
swanOne56.2nativenativenative81 %93 %
Piccolo 27B75.8nativenativenative91 %98 %
Mercury Decide65.9nativenativenative90 %90 %
Winnow-12B Q859.5nativenativenative86 %92 %
cribl-decision-1.0-27B58.6nativenativenative84 %97 %
Bobcat 1.368.1nativenativenative92 %94 %
JPT-35B-A3B51.4nativenativenative74 %94 %
Chon-1.0-12B-Fast60.3nativenativenative100 %78 %
baselabs elephant-155.8nativenativenative78 %91 %
metask-jev-rain-31B-trained70.7nativenativenative100 %88 %
deck-31B73.0nativenativenative96 %94 %
Cygnet54.8nativenativenative81 %92 %
metask-jev-rain-12B-trained66.5nativenativenative100 %86 %
DOPEST 1.0 31B73.9nativenativenative100 %90 %
Eikos-27B64.4nativenativenative81 %98 %
Jev 1.13.063.6nativenativenative79 %98 %
Jeffan-31b-171071.6nativenativenative87 %94 %
Kev 27B64.2nativenativenative80 %97 %
Gemini Decision 167.4nativenativenative83 %98 %
Jev-Omni55.5nativenativenative74 %94 %
Celeris-1 Decision58.9nativenativenative79 %93 %
GEV-26B-Decide49.8below gatenativenativenative70 %96 %
decisio v0.8.0 on gemma-4-12B-it52.3nativenativenative82 %91 %
Myles-G 31B73.6nativenativenative100 %89 %
RSI-Jev v6.1-VL 27B61.6nativenativenative83 %91 %
Nagi-ENORMOUS 27B62.4nativenativenative84 %91 %
StartLux-Decision v1.1 27B AutoThink66.7nativenativenative77 %97 %
Jev 1.14.059.9nativenativenative76 %95 %
metask-jev-rain-31B-trained service v372.6nativenativenative100 %89 %
decisio v0.9.0 on gemma-4-12B-it52.3nativenativenative81 %91 %
Bespoke Nimble 9B v356.5nativenativenative87 %89 %
AutoTrust-Decider-Pro-27B56.8nativenativenative79 %95 %
torchcast-decision-12b60.5nativenativenative96 %86 %
Messier One v0.452.0nativenativenative100 %78 %
shisa-de-149.9below gatenativenativenative76 %90 %
Microsoft-Decision-157.2nativenativenative77 %93 %
Clef59.1nativenativenative82 %94 %
wity-167.9nativenativenative91 %94 %
Wald 4B v2.152.0nativenativenative100 %78 %
Hopper 12B trained51.7nativenativenative79 %93 %
Blink v0.3 26B-A4B49.2below gatenativenativenative67 %95 %
Decision 2.0 Vega 27B60.4nativenativenative73 %99 %
René-1 31B FP861.7nativenativenative78 %98 %
AutoJev-27B55.6nativenativenative70 %99 %
OpenJev70.2nativenativenative96 %93 %
Jebadiah 27B57.5nativenativenative74 %98 %
decider chat on Gemma-4-31B-it58.5nativenativenative80 %98 %
attest-31B-v1.065.2nativenativenative97 %88 %
NInfer Qwen3.8-Flash-Next mixed49.0below gatenativenativenative75 %93 %
Metask rain 4B51.1nativenativenativenot supported—
Sage 1.463.7nativenativenative91 %91 %
FrontiersMind Lumma-fev-27B55.5nativenativenative73 %93 %
Respan Span-0260.9nativenativenative90 %87 %
NInfer Qwen3.8-27B NVFP451.8nativenativenative82 %89 %
SPX-CD Flash58.4nativenativenative81 %94 %
decider-35b-a3b48.7below gatenativenativenative74 %91 %
Diffusion Jev54.9nativenativenative93 %85 %
SPX-CD-Omni48.8below gatenativenativenative68 %93 %
OpenAI Decisions56.9—————
Instinct47.8below gate—————
HS1 0.1.252.9nativenativenative83 %88 %
KnowLine 31B-Gen177.4nativenativenative91 %97 %
Amazon Strands Decider 26B-A4B46.7below gatenativenativenative72 %91 %
Turion Jev 31B70.9nativenativenative87 %95 %
KnowLine 4B-Gen5.546.4below gatenativenativenative70 %89 %
JevOne46.0below gatenativenativenative62 %96 %
Seb-9B-v246.1below gatenativenativenative70 %91 %
pplx-decider-flash-v1.245.5below gatenativenativenative79 %87 %
APUS-OpenJev-v1-9B47.2below gatenativenativenative88 %85 %
Aplomb 145.2below gatenativenativenative68 %93 %
JADE47.9below gatenativenativenative94 %91 %
Tern-35B-A3B44.1below gatenativenativenative63 %96 %
AJev lm6q71.5nativenativenative87 %98 %
NVIDIA Nemotron 3.5 Super VL57.5nativenativenative100 %88 %
Arcbright Elf 1 12B43.8below gatenativenativenative63 %97 %
Plumb-4B42.8below gatenativenativenative71 %91 %
SPX-CD Pro64.6nativenativenative84 %95 %
Messier One v0.242.4below gatenativenativenative85 %85 %
spark-s1-4b-v643.2below gatenativenativenative84 %83 %
Bev / Bonsai 27B43.8below gatenativenativenative70 %91 %
Clef-Flash42.2below gatenativenativenative68 %93 %
Quyet-1.0-Medium41.7below gatenativenativenative76 %84 %
Open-Jev 9B43.9below gatenativenativenative81 %86 %
LitJev43.3below gatenativenativenative66 %98 %
Bespoke Nimble 9B43.1below gatenativenativenative77 %87 %
SimpleJev Qwen3.8-27B57.9nativenativenative78 %96 %
Qwen3.5-9B Jev-like data-mix v241.9below gatenativenativenative71 %91 %
cribl-decision-1.0-9B40.9below gatenativenativenative76 %93 %
lev41.1below gatenativenativenative84 %83 %
Strike-142.5below gatenativenativenative71 %97 %
JPT-4B40.8below gatenativenativenative66 %90 %
jev-local41.9below gatenativenativenative85 %82 %
Aleria Judge 1.1.168.0nativenativenative94 %93 %
Saif 4B40.1below gatenativenativenative71 %94 %
Bespoke Nimble-v441.4below gatenativenativenative57 %96 %
decider-4b v240.1below gatenativenativenative64 %92 %
Jeffan-4B-201040.1below gatenativenativenative63 %91 %
ClassOne Qwen 3.5 9B42.2below gatenativenativenative97 %79 %
Vansa-3.440.9below gate—————
Wald 4B41.0below gatenativenativenativenot supported—
Granite Switch 4.1 30B44.6below gatenativenativenative77 %89 %
Open-Jev 27B v1.152.5nativenativenative67 %94 %
GPT-6 Luna95.7—————
GLiClass v4 9B39.6below gatenativenativenative70 %89 %
Autoloops – Gemma 4 31B IT58.1—————
metask-jev-4b39.2below gatenativenativenative75 %88 %
cribl-decision-238.8below gatenativenativenative60 %98 %
Kahn1 4B39.7below gatenativenativenative70 %87 %
JevK5 v0.338.5below gatenativenativenative66 %95 %
GPT-6 Luna96.9—————
deck-4B v1.038.6below gatenativenativenative68 %93 %
Seb-9B38.3below gatenativenativenative57 %94 %
Decision-4B37.9below gatenativenativenative72 %88 %
decisor-4b39.6below gatenativenativenativenot supported—
janus 4B37.6below gatenativenativenative63 %94 %
Jebadiah 9B v2.137.2below gatenativenativenative60 %93 %
JevK5 v0.2.036.3below gatenativenativenative62 %93 %
Amos 1 medium38.1below gatenativenativenative56 %97 %
Decision 2.0 Lux 9B35.8below gatenativenativenative53 %94 %
Verdin-4B35.4below gatenativenativenative65 %86 %
TypeCastLM 1.4.035.4below gatenativenativenative77 %81 %
blink-4b35.0below gatenativenativenative60 %90 %
Fastino GLiDE66.1—————
Imajev-9B36.7below gatenativenativenative54 %93 %
jiwo 4b34.8below gatenativenativenative63 %90 %
Imajev-4B34.6below gatenativenativenative52 %96 %
Uniphore-4B-Decider34.2below gatenativenativenative72 %82 %
Malkuth-4B33.8below gatenativenativenative65 %92 %
Hopper33.6below gatenativenativenative57 %94 %
SimpleJev Qwen3.6-35B-A3B56.3—————
Decisio v0.9.0 on Qwen3.6-35B-A3B34.0below gatenativenativenative54 %95 %
Clef-omni34.5below gatenativenativenative65 %86 %
JEB-35B-A3B33.2below gatenativenativenative54 %94 %
jqv33.9below gatenativenativenative66 %94 %
CoCo-Decision-4B32.7below gatenativenativenative56 %96 %
Decision 4B v1.232.4below gatenativenativenative59 %94 %
Manchego v2.132.5below gatenativenativenative62 %90 %
Standard One 3B SH v434.4below gatenativenativenative67 %84 %
Sors-2B31.4below gatenativenativenative68 %89 %
Jev-Style-2B31.0below gatenativenativenative49 %91 %
Raw Qwen3 4B Instruct 2507 direct logits34.0below gatenativenativenative92 %73 %
Standard One 3B SH32.8below gatenativenativenative63 %84 %
GPT-5.6 Luna93.0—————
Jebadiah 9B v230.6below gatenativenativenative52 %93 %
reflex 4B30.8below gatenativenativenative58 %92 %
typecastlm29.8below gatenativenativenative57 %90 %
Amos 1 large68.2nativenativenative85 %97 %
iota-decide-4b29.4below gatenativenativenative60 %84 %
Decision 4B v1.129.1below gatenativenativenative57 %94 %
Standard One 8B SH34.5below gatenativenativenative61 %88 %
kev 8B29.3below gatenativenativenative77 %81 %
Compass 0.2.028.9below gatenativenativenative25 %97 %
Granite Switch 4.1 8B29.3below gatenativenativenative62 %88 %
pjev-2b28.8below gatenativenativenative59 %88 %
Instinct Dual 4B28.3below gate—————
ArseneLupin v1.128.0below gatenativenativenative42 %94 %
Gemini 3.1 Flash-Lite58.6—————
Standard One 8B32.8below gatenativenativenative68 %90 %
Arbiter-4B27.4below gatenativenativenative65 %84 %
system-one-open27.9below gate—————
Decider-4B27.1below gatenativenativenative54 %90 %
SemIf, formerly OpenJev27.1below gatenativenativenative62 %90 %
RYOTIDE-Qwen927.6below gatenativenativenativenot supported—
ClassOne Gemma 4 E4B28.2below gatenativenativenative93 %74 %
JEV Qwen3.5-9B Base NVFP429.3below gatenativenativenative53 %89 %
DecisionNode Edge 0.9.226.8below gatenativenativenative62 %83 %
openjev-sglang38.8below gate—————
ClassOne Qwen 3.5 4B27.3below gatenativenativenative86 %71 %
system-one29.3below gatenativenativenative92 %73 %
Jez-large-1.0.025.6below gatenativenativenative34 %98 %
Chon-1.0-12B27.3below gatenativenativenative100 %90 %
RYOTIDE-Qwen25.2below gatenativenativenative57 %89 %
Noma25.1below gatenativenativenative51 %88 %
Amber 4B24.8below gatenativenativenot supported60 %92 %
Raw Qwen3 8B direct logits26.5below gatenativenativenative92 %71 %
local-jev Qwen3.5-4B24.1below gatenativenativenative35 %99 %
baselabs joey-123.5below gatenativenativenative47 %87 %
OpenSourceJev23.2below gatenativenativenative49 %95 %
Decision 2B22.9below gatenativenativenative39 %94 %
open-alternative-jev22.3below gatenativenativenative60 %87 %
Krino-2B22.0below gatenativenativenative83 %71 %
Liquid AI d1-3B22.1below gatenativenativenative49 %83 %
Open-Jev 2B21.8below gatenativenativenative67 %75 %
classify-121.8below gatenativenativenative9 %100 %
Imajev-2B21.2below gatenativenativenative38 %84 %
decider-2b20.7below gatenativenativenative68 %78 %
D1A on Gemma 4 E4B19.9below gatenativenativenative46 %84 %
Nemotron Diffusion 8B20.4below gatenativenativenative61 %77 %
DeepSeek V4.1 Flash84.5—————
kev 4B19.5below gatenativenativenative67 %82 %
cribl-decision-2-mini18.6below gatenativenativenative28 %86 %
DecisionNode Edge Mini 0.9.217.6below gatenativenativenative47 %79 %
Malkuth-2B15.9below gatenativenativenative53 %84 %
Qwen3-Reranker-4B16.3below gatenativenativenative25 %75 %
jiwo 0.8b15.9below gatenativenativenative36 %76 %
ZeroEntropy zerank-215.6below gatenativenativenative14 %87 %
Vega 4B15.7below gatenativenativenative69 %70 %
EXAONE-4.0-1.2B-JEV v0.314.8below gatenativenativenative40 %85 %
SimpleJev Qwen3.8-27B52.9—————
Raw Phi-4 mini direct logits13.7below gatenativenativenative50 %61 %
decision-machine-113.1below gate—————
SimpleJev12.7below gatenativenativenative100 %58 %
Jeetu Decider 2B10.2below gatenativenativenative38 %76 %
JevAct10.1below gate—————
D1A on Gemma 4 E2B8.9below gatenativenativenative28 %78 %
smalljev semantic-v98.7below gatenativenativenative23 %73 %
Gutsy 0.8B v0.38.6below gatenativenativenative29 %74 %
Vega-U8.5below gatenativenativenative33 %76 %
PurpleMIST-Mini-1.08.4below gatenativenativenative9 %88 %
kev 0.6B7.6below gatenativenativenative43 %69 %
jul fast7.5below gatenativenativenative30 %89 %
OpenDecision7.4below gatenativenativenative51 %68 %
ClassOne Gemma 4 E2B7.8below gatenativenativenative83 %59 %
Fastino GLiNER-2.5-Decide7.0below gateconfidenceconfidenceconfidence100 %54 %
Deem 0.8B v16.5below gatenativenativenative77 %59 %
Decision Fast6.1below gatenativenativenative14 %90 %
Quyet-1.0-Small-EN6.1below gatenativenativenative28 %64 %
ClassOne Qwen 3.5 2B5.8below gatenativenativenative87 %62 %
Laya typed-decisions5.5below gatenativenativenative1 %100 %
Vega-U5.3below gatenativenativenative20 %53 %
WaterSheep5.3below gatenativenativenative37 %69 %
openJev Verdict 1.45.0below gatenativenativenative0 %—
Raw Qwen3 0.6B direct logits5.6below gatenativenativenative100 %42 %
lev-350m4.8below gatenativenativenative8 %69 %
Raw Qwen3 1.7B direct logits5.1below gatenativenativenative100 %42 %
kev 0.5B4.6below gatenativenativenative26 %56 %
Jez-small-1.0.04.6below gatenativenativenative11 %85 %
openJev Verdict4.1below gatenativenativenative33 %60 %
Quyet-1.0-Small4.0below gatenativenativenative29 %71 %
watt-flash-0.13.8below gatenativenativenative1 %60 %
Quyet-1.0-Tiny3.7below gatenativenativenative20 %64 %
Phocinae-Largha-150M-v13.6below gatenativenativenative3 %55 %
Liquid AI d1-omni-600M3.3below gatenativenativenative27 %56 %
Vega 0.8B3.1below gatenativenativenative32 %63 %
open-jev-deberta-v3-large2.8below gatenativenativenative15 %58 %
Julia-12.4below gatenativenativenative73 %46 %
verdict-small2.3below gatenativenativenative45 %62 %
Tacet Sonata2.2below gatenativenativenative32 %56 %
Laya1.8below gatenativenativenative19 %61 %
Mixedbread mxbai-rerank-base-v21.1below gatenativenativenative0 %—
Laya multilingual0.3below gatenativenativenative72 %50 %
Certo v10.2below gatenativenativenative1 %25 %
CLM-8B0.1below gatenativenativenative51 %69 %
BAAI bge-reranker-v2-m30.1below gatenativenativenative0 %—
Alibaba GTE Reranker ModernBERT-base0.0below gatenativenativenative0 %—
Mirror0.0below gatenativenativenative92 %60 %
Open Jev JSON Canvas61.2labellabellabel100 %85 %
Qwen3.8 27B33.4below gate—————
All data (357 systems)

Not yet measured on v1.6 · 0 systems with a dated carried score

v1.5 These rows were measured before the 10 s rule (v1.5 method), not re-scored: their scores are not comparable with the re-scored rows above.

Every ranked system of the live v1.5.7 board that is not measured on the v1.6.0 pool keeps its last published score, marked with the release that first published that measurement and that release's publication day. Carried rows are listed separately and never ranked together with v1.6-measured rows. v1.5 protocol (1,624 decisions: 904 open + 720 sealed); scores are on the v1.5 scale and are not comparable with v1.6-measured rows.

Dates: . The date is the publication day of the release that first published the measurement, not a per-model measurement timestamp.

Carried JevBench v1.5.x results, not ranked with v1.6 measurements
SystemMeasured onCapability (v1.5 scale)v1.5 CompositeCost / 1,000Median latency

* Self-hosted rows: the first value is the p50 we measured on our own hardware over loopback (on-GPU, or on-CPU where marked, no network). It is not directly comparable with API latency, which we time end to end from our measuring host, network included. “est.” is the est. incl. network (2x + 150 ms) that the ranking and the latency cap use for every self-hosted row. Model authors can ask us to time a hosted endpoint instead (on request).

Method · v1.8.11

System types (colours)

  • Jev — reference (TypeSafe, closed) — The closed system JevBench is named after, shown as the reference.
  • Closed API (weights not public) — Available through a hosted API; the weights cannot be downloaded.
  • Open weights · LLM decoder — An autoregressive language model with public weights, including fine-tunes, merges and Jev rebuilds.
  • Open weights · diffusion LM — A language model that generates by iterative denoising instead of token by token.
  • Open weights · encoder / classifier — BERT-style encoders, NLI zero-shot classifiers and GLiNER-type models.
  • Open weights · reranker — A cross-encoder or LLM reranker that scores options against the input.
  • Base model control (no decision fine-tune, raw logits) — An official open checkpoint without any decision fine-tune, read out from raw logits, used as a floor.
  • System (router / cascade / ensemble) — Several models combined at inference time.

Colours show architecture only; they never change a score or rank. Each class is assigned from cited evidence (config.json, model card, provider docs, or our own run record for hosted APIs).

Self-hosted rows use S 1,200 + P 300 + L1 (354 items) + L2 (33 items) + L3 (1,118 items), where answered. API rows re-run on A4/A5 use their 300/300-item sealed subset instead of S. Every row’s tag lists the pools it actually answered. L3 is a sealed language supplement (1,118 items), drawn 2026-10-07. Every language has at least 60 items in P ∪ L1 ∪ L2 ∪ L3. Header counts show the full S + P + L1 + L2 + L3 pool. Headline scores are unchanged; these raw cells are unequated, scored for language/category views only and outside the Composite. L3 items were written natively by Claude Sonnet 5.5, each solved blind and language-checked by GPT-6.1 Sol, with gold kept only when both agree or a second review confirms; no gold comes from Jev or any measured API. L3 is API-facing by design and is excluded from future headline draws. Rows with unfinished runs retain their actual coverage tags. L3 items were reviewed by an OpenAI model, so the L3-based language and category cells of OpenAI rows (OpenAI Decisions, GPT-6 Luna, GPT-5.6 Luna) carry that exposure; headline scores do not use L3. C1 adds English items for thin radar categories (everyday language, safety and the other use case). L4 adds 20 new Hindi items for Fastino GLiNER 2.5 Decide only. They were authored with OpenAI and independently solved and language-reviewed with Anthropic before sealing. Seventeen L4 requests succeeded; three HTTP 400 failures remain in the scored observations and do not count as completed coverage. L4 is outside headline and category scores and the common header counts. L5 adds 60 new items for OpenJev DeBERTa v3 Large only: 20 each in Arabic, Hindi and Greek. They were authored with OpenAI and independently solved and language-reviewed with Anthropic before sealing, then checked against the original 512-token native context before keeper selection. All 60 native requests succeeded. Previous failures remain scored and do not count as completed coverage. L5 is outside headline and category scores and the common header counts.

Method amendments

  • 5 Oct 2026, v1.6.1 (published rule): hosted/API systems now answer the same full item set as self-hosted systems (1,200 sealed + 300 public). Their earlier answers on the API subsets were reused and only the missing sealed items were sent; every item was logged in the exposure ledger before it was sent. API rows are no longer equated. Consequence: every v1.6.0 sealed item has now been seen by at least two providers, so the whole v1.6.0 sealed draw is retired for future releases; v1.6.2 onward draws fresh items from the reserve.
  • 5 Oct 2026, v1.6.1 cost rule (published rule): cost per 1,000 decisions is now measured on one common item set for every system: all items except the 23 very long items that lie outside the Jev reference's accepted input range. Before, a system that processed those items paid for their tokens while a system that refused them did not. Refusals still count against Intelligence exactly as before. Hosted APIs with a per-token tariff are priced from their measured tokens on this common set; carried per-decision prices of self-hosted systems are unchanged. Effect: Sage 1.3.0 USD 0.0247 per 1,000 decisions on the common set (0.0650 on all 1,500 items); Jev 1.13.0 0.0323 (unchanged).

Cost rule, generalised in v1.6.30: On every draw, very long items that a ranked row refuses because they exceed its context length are left out of cost for all rows: cost per 1,000 decisions is computed on the same remaining items for every row. A refused item still counts as a wrong answer for Intelligence. First published as the v1.6.1 cost rule (5 Oct 2026; 23 items on that draw). From v1.6.30 it applies to draw v1.6-regular-20261012: the 15 items the Jev reference refused with max_tokens_exceeded are left out and 1,485 remain. Rows of draw v1.6-regular-20261010-a2 still show the all-items cost they were published with until a following release re-costs them under this rule. The 23 items above are the set of the v1.6.1 draw.

The v1.6.1 amendments above say hosted APIs are no longer equated. Exception since v1.7.7: the 17 API rows re-run on A4 ∪ P (600 items) are equated (+2.57 Intelligence, +3.35 Calibration); see “Full API re-run (A4, v1.7.7)” under Open weights and API offerings. Full-set API rows (Jev, Sage, wity-1, Fastino GLiNER-2.5-Decide, d1 (Liquid AI)) are not equated. OpenAI Decisions (gpt-6-luna) ran later on the fresh sealed API set A5 ∪ P (600 items; A4 had already been sent to that provider) and is equated the same way with A5's own offsets (+5.78 Intelligence, +6.15 Calibration, same pool).

Rotating item sets

Each release draws fresh sealed decisions from a larger reserve. Self-hosted open-weights models (run offline on our own GPU pods or Sandy) answer S and P; externally hosted models answer only the API subset A and P.

SetItemsChoiceNoulScoreAnswered by
S · Sealed release draw1,200600300300self-hosted systems (run offline on our own GPU pods or Sandy) and, from v1.6.1, externally hosted APIs
A · API subset (part of S; used by the v1.6.0 hosted-API measurements)3001507575every system (inside S since v1.6.1); retired for future draws
P · Public set3001507575every system

Not in the table above (frozen release data): API set A4 · 300 never-used sealed items, answered together with P by the 18 API endpoints re-run on 6 Oct 2026 (the 17 ranked offerings and the classifier.dev wrapper); retired after that run.

  • Self-hosted systems: 1,500 items (S 1,200 + P 300). Hosted APIs: 1500 items (A 300 + P 300). Exception since v1.7.7: the 17 API rows re-run on A4 ∪ P (600 items) are equated (+2.57 Intelligence, +3.35 Calibration); see “Full API re-run (A4, v1.7.7)” under Open weights and API offerings. Full-set API rows (Jev, Sage, wity-1, Fastino GLiNER-2.5-Decide, d1 (Liquid AI)) are not equated. OpenAI Decisions (gpt-6-luna) ran later on the fresh sealed API set A5 ∪ P (600 items; A4 had already been sent to that provider) and is equated the same way with A5's own offsets (+5.78 Intelligence, +6.15 Calibration, same pool).
  • A sealed item is scored in at most three releases, then retired. An item used in one release is not drawn again in the next. If coverage minimums cannot be met, the release waits for newly reviewed items.
  • The selection seed is committed (SHA-256) before any inference, and the draw is a deterministic function of policy, seed and item id.

API-exposure rule

  • An external exposure is any sealed item sent to an endpoint we do not control: closed APIs, and open-weights models reached through third-party hosts, routers or a submitter's endpoint. Timeouts count as exposure.
  • Hosted models receive only the release's API subset A plus the public set P, and are re-measured at most once every three refresh releases unless a verified new model version ships. Between measurements they keep their last score with its measurement date.
  • Every externally sent sealed item is logged before dispatch. An item exposed to a provider is never scored again for that provider; once two different providers have received it, it retires for everyone. In v1.6.0 Jev 1.13.0 and Fastino GLiNER-2.5-Decide both received the same A, so A retires globally.

Public-versus-sealed gap penalty

Intelligence is half public, half sealed. A system whose public score exceeds its sealed score by more than the field-median gap (G_med = 2.6 points) plus 8 points loses one Intelligence point per excess point. Hosted APIs are compared on P versus A against the same self-hosted systems' P-versus-A gap (-1.2 points).

Comparable scores for hosted APIs

Median over complete full-coverage self-hosted systems of metric(S u P) - metric(A u P), per metric (Intelligence, Calibration); API value = clip(raw_A + offset). Offsets in this run: Intelligence -2.21, Calibration +2.34. Historical v1.6.0 equating offsets, kept for the record. Hosted APIs that answered the full set (v1.6.1) are not equated. Category and language values stay raw and unequated.

Headline and Composite

Capability = mean(Intelligence, Calibration) for systems within twice the Jev 1.13.0 cost and median latency (the official caps; the sliders change only your view). The Composite (option A) is the equal-weight harmonic mean of Intelligence, Calibration, Speed and Cost with the v1.5 low-axis gates; it remains secondary. Request types Choice, Noul and Score weigh equally; tiers weigh easy 0.1, standard 0.2, hard 0.4, judge 0.3.

Latency. API models are timed end to end from our measuring host, network included. Self-hosted models are timed on our own hardware over loopback (on-GPU, no network), so their measured value is not directly comparable with API latency. The ranking and the latency cap use the same estimate for every self-hosted row: 2 × measured p50 + 150 ms (est. incl. network (2x + 150 ms)). Tables show the measured value first, tagged “on-GPU, no network”, and the estimate second. A model author can provide a hosted endpoint; we then time it exactly like an API, network included, with no assumption (on request: info@productivity-boost.com, or note it in your submission). From JevBench v1.7.0 (method v1.7), a decision that takes longer than 10 seconds (measured time, not the estimate) counts as a wrong answer and as zero calibration for that item; tables show the measured p95 and how many decisions were over 10 s.

Jev-class caps use a fixed reference: Jev 1.13.0 as measured in v1.5 (p50 0.62 s, USD 0.0323 per 1,000 decisions); caps = 2x (1.23 s, USD 0.0646). Jev's own v1.6 p50 is 0.24 s.

Costs of v1.6-measured systems carry each system's published v1.5.4 cost per 1,000 decisions (pricing rules unchanged; v1.6 item lengths differ) unless the row says otherwise; Fastino's is an estimate from its published tariff and measured tokens.

Noul decisiveness and Score baseline (addendum B)

Scored with method option B (scorer setting O1S), selected on 3 Oct 2026 after the v1.6 scores were known and disclosed as a post-results change. Each split × type competence is clipped at 0 before the type weighting, so a type answered no better than chance counts as chance instead of negative. The Score chance baseline is the error of always predicting the mid-scale level, so a flat know-nothing distribution earns about 0. A Score cell whose golds all sit at mid-scale keeps the v1.5 random-level baseline; this only occurs in small breakdown and bootstrap cells. Calibration is unchanged. This run: G_med = 2.57 points; hosted-API offsets Intelligence -2.21, Calibration +2.34.

Failed requests and very long items

23 items of this draw are very long (about 77,000 to 81,000 input tokens). Systems with a shorter context window refuse them; Jev-Omni runs out of GPU memory on them on its listed RTX 6000 (48 GB) recipe; decider-4b v2's server truncates them to about 32,800 tokens and answers; Plumb-4B reads them in full. Each outcome is the system's own recipe and is scored as such. A failed, refused or unparseable answer counts as wrong for Intelligence and stays in the denominator; it does not enter Calibration (the v1.5 rule, applied to every system). Failed answers per system: AJev lm6q 15/1500 · Aleria Judge 1.1.1 15/1500 · allec-oavg 5/1125 · Amazon Strands Decider 26B-A4B 15/1500 · Amber 4B 1/1125 · Amos 1 large 21/1500 · Amos 1 medium 10/1500 · Aplomb 1 5/1500 · APUS-OpenJev-v1-35B-A3B 23/1500 · APUS-OpenJev-v1-9B 23/1500 · Arcbright Elf 1 12B 15/1500 · Arcbright Elf 1 12B (r2) 0/1500 · Arcbright Elf 1 31B 15/1500 · ArseneLupin v1.1 15/1500 · attest-31B-v1.0 15/1500 · AutoJev-27B 23/1500 · basal-omni-1.0-ultra 15/1500 · BB-Qwen3.5-4B-LoRA 0/1500 · Bespoke Nimble 9B v3 23/1500 · Bev / Bonsai 27B 23/1500 · BAAI bge-reranker-v2-m3 0/1500 · blink-4b 2/1500 · Blink v0.3 26B-A4B 29/1500 · Bobcat 1.3 15/1500 · Bobcat Flash 1.2 13/1500 · Bongard-mini 15/1500 · Celeris-1 Decision 0/1500 · Certo v1 0/1500 · Jev-Style-2B 5/1500 · Chon-1.0-12B 540/1500 · Chon-1.0-12B-Fast 0/1500 · Chon-1.0-E4B 0/1500 · classifier.dev 9/600 · ClassOne Gemma 4 E2B 23/1500 · ClassOne Gemma 4 E4B 29/1500 · ClassOne Qwen 3.5 2B 29/1500 · ClassOne Qwen 3.5 4B 29/1500 · ClassOne Qwen 3.5 9B 23/1500 · Clef 0/1500 · Clef-Flash 0/1500 · Clef-omni 0/1500 · CLM-8B 1/1500 · CoCo-Decision-4B 23/1500 · Compass 0.2.0 0/1500 · cribl-decision-1.0-27B 5/1500 · cribl-decision-1.0-9B 0/1500 · cribl-decision-2 5/1500 · cribl-decision-2-mini 5/1500 · Cygnet 23/1500 · D1A on Gemma 4 E2B 15/1500 · D1A on Gemma 4 E4B 15/1500 · decide-1.1 2/1500 · decider-12b v2 23/1500 · decider-12b v1, stock Gemma-4-12B-it 23/1500 · decider-2b 0/1500 · decider-35b-a3b 0/1500 · decider-4b v2 0/1500 · decider chat on Gemma-4-31B-it 23/1500 · Decisio 12B tuned commit 15/1500 · Decisio 12B tuned commit-think 0/1500 · decisio v0.8.0 on gemma-4-12B-it 23/1500 · decisio v0.9.0 on gemma-4-12B-it 23/1500 · decisio v0.8.0 on gemma-4-31B-it 23/1500 · Decisio v0.9.0 on Qwen3.6-35B-A3B 15/1500 · Decision 2.0 Vega 27B 23/1500 · Decision 2B 23/1500 · Decision 4B v1.1 23/1500 · Decision 4B v1.2 23/1500 · Decision Fast 23/1500 · Decision-King v1.1 6/1500 · decision-machine-1 9/600 · Decision 2.0 Lux 9B 15/1500 · DecisionNode Edge 0.9.2 15/1500 · DecisionNode Edge Mini 0.9.2 15/1500 · DecisionNode Flash 5/1500 · DecisionNode 5/1500 · decisor-4b 0/1500 · deck-4B v1.0 23/1500 · deck-31B 23/1500 · Deem 0.8B v1 23/1500 · DeepSeek V4.1 Flash 5/600 · Diffusion Jev 26/1500 · DOPEST 1.0 31B 0/1500 · Drex Max 2/1500 · Eikos-27B 23/1500 · baselabs elephant-1 15/1500 · Decision-4B 23/1500 · EXAONE-4.0-1.2B-JEV v0.3 23/1500 · Fastino GLiDE 9/600 · Fastino GLiDE 1.2 15/1497 · Fastino GLiDE (hosted API, 22 Oct 2026) 19/1497 · Fastino GLiNER-2.5-Decide 31/1500 · Gemini 3.1 Flash-Lite 1/600 · Gemini Decision 1 3/1500 · GEV-26B-Decide 23/1500 · Arbiter-4B 15/1500 · GPT-5.6 Luna 0/600 · GPT-6 Luna (default medium reasoning effort) 4/600 · GPT-6 Luna (low reasoning effort) 0/600 · Granite Switch 4.1 30B 5/1500 · Granite Switch 4.1 8B 5/1500 · Alibaba GTE Reranker ModernBERT-base 0/1500 · Gutsy 0.8B v0.3 29/1500 · H2O-Lightning-31B v1.0.7 15/1500 · H2O-Lightning-4B v1.1 0/1500 · H2O-Lightning-4B v1.4.1 0/1500 · Hopper 22/1500 · Hopper 12B trained 23/1500 · HS1 0.1.2 15/1500 · Imajev-2B 15/1500 · Imajev-4B 23/1500 · Imajev-9B 15/1500 · Instinct 9/600 · Instinct Dual 4B 0/600 · iota-decide-4b 0/1500 · iota-decide-4b-v0.3 0/1500 · iota-decide-4b-v0.4 0/1500 · JADE 392/1500 · janus 4B 23/1500 · JEB-35B-A3B 15/1500 · Jebadiah 27B 24/1500 · Jebadiah 9B v2 15/1500 · Jebadiah 9B v2.1 15/1500 · Jeetu Decider 2B 15/1500 · Jeff-1.0-Large 21/1500 · Jeff-1.1-Large 0/1500 · Jeff-1.1-Medium 0/1500 · Jeff-1.1-Qwen-4B 0/1500 · Jeffan-31b-1710 15/1500 · Jeffan-4B-2010 15/1500 · Jev 1.13.0 23/1500 · Jev 1.14.0 15/1500 · AutoTrust-Decider-Pro-27B 23/1500 · jev-local 23/1500 · Jev-Omni 23/1500 · JEV Qwen3.5-9B Base NVFP4 23/1500 · JevAct 21/600 · JevK5 v0.3 23/1500 · JevK5 v0.2.0 23/1500 · JevOne 0/1500 · jevora-9b-v1 49/1500 · Jez-large-1.0.0 15/1500 · Jez-small-1.0.0 15/1500 · jiwo 0.8b 29/1500 · jiwo 4b 29/1500 · baselabs joey-1 5/1500 · JPT-35B-A3B 23/1500 · JPT-4B 15/1500 · jqv 23/1500 · jul fast 6/1500 · Julia-1 17/1500 · Kahn1 4B 23/1500 · Karpo-Decision-31B 0/0 · kettei-chub 15/1500 · kettei-fat 15/1500 · kev 0.5B 32/1500 · kev 0.6B 30/1500 · Kev 27B 23/1500 · kev 4B 29/1500 · kev 8B 25/1500 · Kev-9B v2 15/1500 · GLiClass v4 9B 15/1500 · KnowLine 26B-A4B-Gen2.5 15/1500 · KnowLine 31B-Gen1 15/1500 · KnowLine 35B-A3B-Gen1.5 0/1500 · KnowLine 4B-Gen5.5 0/1500 · Krino-2B 0/1500 · Autoloops – Gemma 4 31B IT 9/600 · Myles-G 12B 0/1500 · Myles-G 31B 0/1500 · Laya 0/1500 · Laya multilingual 0/1500 · Laya typed-decisions 0/1500 · lev 25/1500 · lev-350m 23/1500 · d1 0/1500 · Liquid AI d1-3B 0/1500 · Liquid AI d1-omni-600M 0/1500 · d1-ultra 15/1500 · LitJev 23/1500 · local-jev Qwen3.5-4B 0/1500 · FrontiersMind Lumma-fev-27B 15/1500 · Lychee-27B 15/1500 · Malkuth-2B 23/1500 · Malkuth-4B 23/1500 · baselabs mammoth-1 15/1500 · Manchego v2.1 23/1500 · Mercury Decide 23/1500 · Messier One v0.2 0/1500 · Messier One v0.4 0/1500 · MetaDecision-31B-Rain 0/1500 · metask-jev-4b 23/1500 · Metask rain 4B 0/1500 · metask-jev-rain-12B 0/1500 · metask-jev-rain-12B-trained 12/1500 · metask-jev-rain-31B-trained 0/1500 · metask-jev-rain-31B-trained service v3 0/1500 · Qwen3.5-0.8B Decision Model 0/1500 · Microsoft-Decision-1 (Azure Foundry) 0/0 · Microsoft-Decision-1 (updated weights, Oct 2026) 4/1493 · Minimodels Decision 1.0 0/0 · Mirror 592/1500 · modernbert-jev v2 0/1500 · Mixedbread mxbai-rerank-base-v2 0/1500 · Nagi-ENORMOUS 27B 15/1500 · Nagi-HUGE 12B 15/1500 · Nemotron Deciffusion 8B v2 0/1500 · Nemotron Diffusion 8B 23/1500 · Nev-2B 15/1500 · Bespoke Nimble 9B 23/1500 · Bespoke Nimble-v4 0/1500 · Bespoke Nimble-v4-max 0/1500 · Nimble-v4-max 16/1500 · NInfer Qwen3.8-27B NVFP4 23/1500 · NInfer Qwen3.8-Flash-Next mixed 23/1500 · Noma 0/1500 · NVIDIA Nemotron 3.5 Super VL 2/1500 · oenogarum 5/1125 · open-alternative-jev 0/1500 · open-jev-deberta-v3-large 73/1500 · Open Jev JSON Canvas 23/1500 · Open-Jev 27B v1.1 23/1500 · Open-Jev 2B 23/1500 · Open-Jev 9B 23/1500 · OpenAI Decisions 2/600 · OpenDecision 0/1500 · openjev-sglang 9/600 · OpenJev 24/1500 · openJev Verdict 0/1500 · openJev Verdict 1.4 0/1500 · OpenSourceJev 23/1500 · Philly Decider v2 2/1500 · Phocinae-Largha-150M-v1 375/1500 · Piccolo 27B 5/1500 · Piccolo 4B 5/1500 · Piccolo 9B 5/1500 · pjev-2b 15/1500 · Plumb-4B 2/1500 · pplx-decider-flash-v1.2 15/1500 · Perplexity Decider v1.1 27B 24/1500 · pplx-decider-v1.2 15/1500 · PurpleMIST-Flash-1.0 15/1500 · PurpleMIST-Mini-1.0 15/1500 · Quyet-1.0-Large 0/1500 · Quyet-1.0-Medium 0/1500 · Quyet-1.0-Small 0/1500 · Quyet-1.0-Small-EN 0/1500 · Quyet-1.0-Tiny 0/1500 · Qwen3-Reranker-4B 0/1500 · Qwen3.8 27B 8/600 · Qwen3.5-9B Jev-like data-mix v2 23/1500 · Raw Phi-4 mini direct logits 23/1500 · Raw Qwen3 0.6B direct logits 23/1500 · Raw Qwen3 1.7B direct logits 23/1500 · Raw Qwen3 4B Instruct 2507 direct logits 23/1500 · Raw Qwen3 8B direct logits 23/1500 · reflex 4B 2/1500 · Reflex Instinct 0.6B 0/1500 · René-1 31B FP8 23/1500 · Respan Span-02 15/1500 · Decider-4B 15/1500 · RSI-Jev v6.1-VL 27B 29/1500 · RYOTIDE-Gemma 15/1500 · RYOTIDE-Qwen 2/1500 · RYOTIDE-Qwen9 21/1500 · Sage 1.3.0 0/1500 · Sage 1.4 30/1500 · Saif 4B 0/1500 · classify-1 14/1500 · SD-JEV-4B-v12.1.0 5/1500 · Seb-9B 15/1500 · Seb-9B-v2 15/1500 · SemIf, formerly OpenJev 23/1500 · shisa-de-1 16/1500 · Sieve-31B-v2 15/1500 · SimpleJev 37/1500 · SimpleJev Qwen3.6-35B-A3B 9/600 · SimpleJev Qwen3.8-27B 9/600 · SimpleJev Qwen3.8-27B (self-hosted, v1.6 pool) 23/1500 · Super Decider 11/1500 · Super Decider (thinking) 4/1500 · smalljev semantic-v9 0/1500 · Sors-2B 5/1500 · spark-s1-4b-v6 0/1500 · SPX-CD Flash 23/1500 · SPX-CD-Omni 23/1500 · SPX-CD Pro 24/1500 · Standard One 3B SH 15/1500 · Standard One 3B SH v4 15/1500 · Standard One 8B SH 29/1500 · Standard One 8B 23/1500 · StartLux-Decision v1.1 27B AutoThink 27/1500 · StartLux-Decision v1.1 27B Instant 15/1500 · Strike-1 5/1500 · SubQ Snap 1.1 0/1500 · SubQ Snap 0/1500 · Surogate Rune 26B-A4B v3 23/1500 · swanOne 23/1500 · system-one-open 0/600 · system-one 23/1500 · Tacet Sonata 3/1500 · TDE general large v0.1 0/1500 · Tern-35B-A3B 15/1500 · torchcast-decision-12b 23/1500 · torchcast-decision-27b 23/1500 · Torchcast Decision 31B v1.0 0/1500 · Torchcast Decision 4B (v1.1) 0/1500 · Torchcast Decision 4B (v1.2) 0/1500 · Torchcast Decision 4B (v1.3) 0/1500 · Trio-Spark v1.3 102/1500 · Turion Jev 31B 15/1500 · typecastlm 0/1500 · TypeCastLM 1.4.0 0/1500 · Umpire 35B-A3B 5/1500 · Uniphore-4B-Decider 15/1500 · Vansa-3.4 9/600 · Vega-U (Qwen3.5-0.8B reader) 15/1500 · Vega-U (Qwen3.5-4B reader) 15/1500 · Vega 0.8B 0/1500 · Vega 4B 29/1500 · Vera Large 15/1500 · verdict-small 0/1500 · Verdin-4B 15/1500 · VisionJEV 1.0.0 5/1500 · vLLM-SR d3 15/1500 · vLLM-SR d3-mini 15/1500 · vLLM-SR d3-turbo 15/1500 · Wald 26B-A4B 0/1500 · Wald 4B 8/1500 · Wald 4B v2.1 0/1500 · WaterSheep 1/1500 · watt-flash-0.1 23/1500 · WeMM-JEV 5/1500 · Winnow-12B Q8 23/1500 · wity-1 (Wity, reasoning auto) 36/1500 · wity-1 (Wity, reasoning always) 36/1500 · wity-1 (Wity, reasoning off) 0/1500 · Xevn-4B 15/1500 · Xor 26B-A4B 23/1500 · ZeroEntropy zerank-2 0/1500 · z3-mini-soup-preview 5/1500.

Calibration basis

Systems that return a full probability distribution are calibrated on all components (top-label error, plus distribution distance for Choice and ranked-probability error for Score). Fastino GLiNER-2.5-Decide returns a single confidence value, so its Calibration is the top-label error only and is not like-for-like with full-distribution systems.

Overnight full re-measure (4–5 Oct 2026)

Every system with a reproducible recipe was re-run on the v1.6.0 pool overnight with the same pinned inputs and scorer (method option B / O1S). This page uses scoring round score-v161-3 (v1.6.1) (2026-10-06 00:39:36 UTC). Only complete runs (1,500 items self-hosted, the full API input for hosted APIs) are ranked; partial runs are never ranked, and systems not yet re-measured keep their dated v1.5.x score in the separate table.

  • Scores are the official v1.6.0 scorer (score_v16.py, method option B / O1S, bootstrap B = 1,000) over complete outputs only; each scoring round is kept separately.
  • Hosted APIs are scored on the same full item set as self-hosted systems (S 1,200 + P 300, v1.6.1) and are not equated.
  • GPU-class deviations: reproducible recipes used the hardware listed per row, including H100 for large fast-lane decoders and RTX 6000 for the baseline; hardware differences remain in measured latency. The standard x2 + 0.15 s self-hosted adjustment is an assumption, not a hardware normalization.
  • Wity auto remains outside the latency cap and stays ranked in Composite A. OFF and ALWAYS are unranked variants of the AUTO main row.
  • 6 further measured candidates await a separate publication decision.
  • Sage 1.3.0 (Levanto Labs) was measured on 5 Oct from Sandy (Helsinki), text only. Cost uses the Levanto list tariff (USD 0.05/M input, USD 10/M output; levanto.ai/pricing, read 5 Oct) over all 600 answered rows: USD 0.0766/1,000 decisions. Superseded by the v1.6.1 cost rule (common item set): USD 0.0247/1,000 decisions, inside the Jev-class cost cap.
  • A fresh Monday Jev 1.13.0 re-check on A3 measured p50 0.239 s and equated Capability 77.5 versus the official 76.5, within the confidence interval; the official v1.6.0 Jev row is retained.

Display note on the overnight text above (frozen release data). Exception since v1.7.7: the 17 API rows re-run on A4 ∪ P (600 items) are equated (+2.57 Intelligence, +3.35 Calibration); see “Full API re-run (A4, v1.7.7)” under Open weights and API offerings. Full-set API rows (Jev, Sage, wity-1, Fastino GLiNER-2.5-Decide, d1 (Liquid AI)) are not equated. OpenAI Decisions (gpt-6-luna) ran later on the fresh sealed API set A5 ∪ P (600 items; A4 had already been sent to that provider) and is equated the same way with A5's own offsets (+5.78 Intelligence, +6.15 Calibration, same pool).

Supplementary API draws A2 and A3

History: hosted APIs were first measured on API subsets (A, then A2 and A3) and equated to the self-hosted scale in v1.6.0. From v1.6.1 they answer the same full set as self-hosted systems, so no equating is applied; the earlier subsets A, A2 and A3 are retired.

Per-model exposure counts (hosted and author-hosted endpoints)

System (provider)v1.6 sealed items sentScored sealed setStatus
classifier-dev-fast (classifier.dev)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
decision-machine-1 (milliseconds.ai)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
deepseek-flash (deepseek)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
fastino-gliner-2-5-decide (fastino)1200S (1,200 sealed) + P (300 public)full set since v1.6.1 (5 Oct 2026); earlier API-subset answers reused, only missing sealed items sent; whole v1.6.0 sealed draw retired
gemini-3.1-flash-lite (google)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
gpt-5.6-luna (openai)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
gpt-6-luna (openai)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
gpt-6-luna-low (openai)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
instinct (zoowork)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
instinct-dual-4b (zoowork)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
jev-1.13.0 (typesafe)1200S (1,200 sealed) + P (300 public)full set since v1.6.1 (5 Oct 2026); earlier API-subset answers reused, only missing sealed items sent; whole v1.6.0 sealed draw retired
jevact (jevact)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
kushal-gemma4-31b-it-autoloops (autoloops)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
openjev-sglang (modal via modal)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
qwen3.8-27b (chutes)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
Sage 1.3.0 (Levanto Labs)1200S (1,200 sealed) + P (300 public)full set since v1.6.1 (5 Oct 2026); earlier API-subset answers reused, only missing sealed items sent; whole v1.6.0 sealed draw retired
simplejev-qwen3.6-35b-a3b (featherless)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
simplejev-qwen3.8-27b (featherless)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
system-one-open (modal via modal)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
vansa-3.4 (vansa)300A4 (retired after this run)measured 6 Oct 2026 on A4 ∪ P (v1.7.7)
wity-1-always (wity via railway)1200S (1,200 sealed) + P (300 public)full set since v1.6.1 (5 Oct 2026); earlier API-subset answers reused, only missing sealed items sent; whole v1.6.0 sealed draw retired
wity-1-auto (wity via railway)1200S (1,200 sealed) + P (300 public)full set since v1.6.1 (5 Oct 2026); earlier API-subset answers reused, only missing sealed items sent; whole v1.6.0 sealed draw retired
wity-1-off (wity via railway)1200S (1,200 sealed) + P (300 public)full set since v1.6.1 (5 Oct 2026); earlier API-subset answers reused, only missing sealed items sent; whole v1.6.0 sealed draw retired

Revision history

  • v1.6.1 · 2026-10-05 · Hosted APIs (fastino-gliner-2-5-decide, jev-1.13.0, sage-1.3.0, wity-1, wity-1-always, wity-1-off) answer the full 1,500-item set instead of the 600-item API subset and are no longer equated; self-hosted rows unchanged; whole v1.6.0 sealed draw retired. Cost per 1,000 decisions is measured on one common item set (all items except the 23 outside the Jev reference input range) for every token-priced row.
  • v1.6.0 · 2026-10-05 · Rotating item draw (1,200 sealed + 300 public); hosted APIs on API subsets equated to the self-hosted scale; /jev-models/v1.6.0 stays available unchanged.

Provenance

Measured on the v1.6 pool (run completion day, UTC): 2026-10-02: AutoJev-27B, BAAI bge-reranker-v2-m3, Certo v1, CLM-8B, Cygnet, decider-2b, decider-35b-a3b, decider-4b v2, Decision 4B v1.1, Decision 4B v1.2, Eikos-27B, Alibaba GTE Reranker ModernBERT-base, Hopper, Imajev-4B, jev-local, Jev-Omni, JEV Qwen3.5-9B Base NVFP4, JevK5 v0.3, JevK5 v0.2.0, JevOne, jqv, kev 0.5B, kev 0.6B, kev 4B, kev 8B, Laya, LitJev, local-jev Qwen3.5-4B, Malkuth-2B, Malkuth-4B, metask-jev-4b, Qwen3.5-0.8B Decision Model, Mirror, Mixedbread mxbai-rerank-base-v2, Bespoke Nimble 9B, NInfer Qwen3.8-27B NVFP4, NInfer Qwen3.8-Flash-Next mixed, open-alternative-jev, open-jev-deberta-v3-large, Open Jev JSON Canvas, Open-Jev 2B, Open-Jev 9B, OpenDecision, OpenJev, openJev Verdict, openJev Verdict 1.4, OpenSourceJev, Plumb-4B, Qwen3-Reranker-4B, Qwen3.5-9B Jev-like data-mix v2, Raw Phi-4 mini direct logits, Raw Qwen3 0.6B direct logits, Raw Qwen3 1.7B direct logits, Raw Qwen3 4B Instruct 2507 direct logits, Raw Qwen3 8B direct logits, reflex 4B, SemIf, formerly OpenJev, SimpleJev, smalljev semantic-v9, Standard One 8B, swanOne, system-one, typecastlm, verdict-small, Winnow-12B Q8, ZeroEntropy zerank-2 · 2026-10-04: Bev / Bonsai 27B, Clef, Clef-Flash, decider-12b v2, decider-12b v1, stock Gemma-4-12B-it, Deem 0.8B v1, Decision-4B, H2O-Lightning-4B v1.1, janus 4B, Laya multilingual, Laya typed-decisions, lev, Manchego v2.1, Nemotron Diffusion 8B, Open-Jev 27B v1.1, Quyet-1.0-Large, Quyet-1.0-Medium, Quyet-1.0-Small, Quyet-1.0-Small-EN, Quyet-1.0-Tiny, torchcast-decision-12b · 2026-10-05: Decision 2B, Decision Fast, deck-4B v1.0, deck-31B, Fastino GLiNER-2.5-Decide, Jev 1.13.0, lev-350m, Sage 1.3.0, spark-s1-4b-v6 · 2026-10-06: Aplomb 1, Bespoke Nimble 9B v3, classifier.dev, decider chat on Gemma-4-31B-it, decisio v0.8.0 on gemma-4-12B-it, decisio v0.8.0 on gemma-4-31B-it, Decision 2.0 Vega 27B, decision-machine-1, DeepSeek V4.1 Flash, Diffusion Jev, Fastino GLiDE, Gemini 3.1 Flash-Lite, GEV-26B-Decide, GPT-5.6 Luna, GPT-6 Luna (default medium reasoning effort), GPT-6 Luna (low reasoning effort), Hopper 12B trained, Instinct, Instinct Dual 4B, JADE, Jebadiah 27B, AutoTrust-Decider-Pro-27B, JevAct, JPT-35B-A3B, Kev 27B, Autoloops – Gemma 4 31B IT, d1, OpenAI Decisions, openjev-sglang, Perplexity Decider v1.1 27B, Qwen3.8 27B, SimpleJev Qwen3.6-35B-A3B, SimpleJev Qwen3.8-27B, SimpleJev Qwen3.8-27B (self-hosted, v1.6 pool), SPX-CD Flash, SPX-CD-Omni, SPX-CD Pro, Surogate Rune 26B-A4B v3, system-one-open, torchcast-decision-27b, Vansa-3.4, wity-1 (Wity, reasoning auto), wity-1 (Wity, reasoning always), wity-1 (Wity, reasoning off), Xor 26B-A4B · 2026-10-07: APUS-OpenJev-v1-35B-A3B, APUS-OpenJev-v1-9B, BB-Qwen3.5-4B-LoRA, Bobcat Flash 1.2, ClassOne Gemma 4 E2B, ClassOne Qwen 3.5 9B, CoCo-Decision-4B, decisio v0.9.0 on gemma-4-12B-it, EXAONE-4.0-1.2B-JEV v0.3, jul fast, Kahn1 4B, Mercury Decide, Messier One v0.2, René-1 31B FP8, Tacet Sonata, TypeCastLM 1.4.0, WaterSheep, watt-flash-0.1 · 2026-10-09: decisor-4b, Jeff-1.0-Large, Metask rain 4B, metask-jev-rain-12B, Wald 4B · 2026-10-10: Blink v0.3 26B-A4B, ClassOne Gemma 4 E4B, ClassOne Qwen 3.5 2B, ClassOne Qwen 3.5 4B, Clef-omni, Gutsy 0.8B v0.3, jiwo 0.8b, jiwo 4b, Liquid AI d1-3B, Liquid AI d1-omni-600M, Microsoft-Decision-1 (Azure Foundry), modernbert-jev v2, RSI-Jev v6.1-VL 27B, RYOTIDE-Qwen9, Standard One 8B SH, Vega 0.8B, Vega 4B, Wald 4B v2.1 · 2026-10-12: attest-31B-v1.0, Krino-2B, metask-jev-rain-12B-trained, metask-jev-rain-31B-trained, Standard One 3B SH, Torchcast Decision 4B (v1.1), Uniphore-4B-Decider, vLLM-SR d3 · 2026-10-13: Amos 1 large, Amos 1 medium, Julia-1, Lychee-27B, metask-jev-rain-31B-trained service v3, Phocinae-Largha-150M-v1, Reflex Instinct 0.6B, Sage 1.4, SubQ Snap · 2026-10-14: Messier One v0.4, Bespoke Nimble-v4, Bespoke Nimble-v4-max, Respan Span-02 · 2026-10-15: d1-ultra, Nemotron Deciffusion 8B v2, Nev-2B, Trio-Spark v1.3 · 2026-10-16: ArseneLupin v1.1, blink-4b, Bongard-mini, Celeris-1 Decision, Compass 0.2.0, Decision 2.0 Lux 9B, Imajev-2B, Imajev-9B, JEB-35B-A3B, Jebadiah 9B v2, Jebadiah 9B v2.1, Jev 1.14.0, jevora-9b-v1, JPT-4B, kettei-chub, kettei-fat, Kev-9B v2, Nagi-ENORMOUS 27B, Nagi-HUGE 12B, Noma, RYOTIDE-Gemma, RYOTIDE-Qwen, shisa-de-1, StartLux-Decision v1.1 27B AutoThink, StartLux-Decision v1.1 27B Instant, TDE general large v0.1 · 2026-10-17: basal-omni-1.0-ultra, PurpleMIST-Flash-1.0 · 2026-10-18: Torchcast Decision 4B (v1.2), Vega-U (Qwen3.5-0.8B reader) · 2026-10-19: Chon-1.0-12B, pplx-decider-flash-v1.2, vLLM-SR d3-turbo · 2026-10-20: Arcbright Elf 1 12B, pplx-decider-v1.2, Seb-9B · 2026-10-21: Drex Max, Arbiter-4B, HS1 0.1.2, iota-decide-4b, Myles-G 12B, Myles-G 31B, Nimble-v4-max, Decider-4B, Super Decider, Super Decider (thinking) · 2026-10-22: AJev lm6q, Aleria Judge 1.1.1, Arcbright Elf 1 31B, baselabs elephant-1, Fastino GLiDE (hosted API, 22 Oct 2026), H2O-Lightning-31B v1.0.7, Jez-large-1.0.0, Jez-small-1.0.0, FrontiersMind Lumma-fev-27B, baselabs mammoth-1, pjev-2b, PurpleMIST-Mini-1.0, SubQ Snap 1.1, Tern-35B-A3B, Verdin-4B · 2026-10-23: Bobcat 1.3, Fastino GLiDE 1.2, KnowLine 26B-A4B-Gen2.5, KnowLine 35B-A3B-Gen1.5, KnowLine 4B-Gen5.5, Seb-9B-v2, Standard One 3B SH v4, Torchcast Decision 4B (v1.3), Vega-U (Qwen3.5-4B reader), Wald 26B-A4B · 2026-10-24: Amazon Strands Decider 26B-A4B, DecisionNode Edge 0.9.2, DecisionNode Edge Mini 0.9.2, Jeffan-31b-1710, Jeffan-4B-2010, KnowLine 31B-Gen1, classify-1, Turion Jev 31B, Vera Large, vLLM-SR d3-mini, Xevn-4B · 2026-10-25: Chon-1.0-12B-Fast, Chon-1.0-E4B, Decisio 12B tuned commit, DOPEST 1.0 31B, Jeetu Decider 2B, Sieve-31B-v2 · 2026-10-26: D1A on Gemma 4 E2B, D1A on Gemma 4 E4B, Decisio v0.9.0 on Qwen3.6-35B-A3B, iota-decide-4b-v0.3, Jeff-1.1-Medium, GLiClass v4 9B, Philly Decider v2 · 2026-10-27: Arcbright Elf 1 12B (r2), Decisio 12B tuned commit-think, H2O-Lightning-4B v1.4.1, Jeff-1.1-Large, Jeff-1.1-Qwen-4B, MetaDecision-31B-Rain, Torchcast Decision 31B v1.0 · 2026-10-28: Jev-Style-2B, DecisionNode Flash, DecisionNode, Microsoft-Decision-1 (updated weights, Oct 2026), Strike-1 · 2026-10-29: cribl-decision-2, cribl-decision-2-mini, decide-1.1, Decision-King v1.1, iota-decide-4b-v0.4, baselabs joey-1, Saif 4B, Umpire 35B-A3B, VisionJEV 1.0.0, WeMM-JEV · 2026-10-30: allec-oavg, Amber 4B, cribl-decision-1.0-9B, oenogarum, Piccolo 27B, Piccolo 4B, Piccolo 9B, Sors-2B · 2026-10-31: cribl-decision-1.0-27B, Granite Switch 4.1 30B, Granite Switch 4.1 8B, z3-mini-soup-preview · 2026-11-03: NVIDIA Nemotron 3.5 Super VL · 2026-11-10: Karpo-Decision-31B, SD-JEV-4B-v12.1.0 · 2026-11-11: Gemini Decision 1 · 2026-11-15: Minimodels Decision 1.0.

Aggregate files: results sha256 3599bbf28797e1da9a50de887ff500a2ffa8643cd2001e9ab3e635e08b592b69 · categories sha256 35eecc0f30bf88f398f30e4ab7b127ef7632070edb1fd267b2d0ba9466655972 · dated carry sha256 271d1fc77a136a04a1eb6deb5480c52d4211eb2b66b2ac8bb31e80867e69003a. Scoring source sha256 b89e7eb86003cad44b9edc1b5b4bcf3f354d579773d859eef6b155997bc4f728. The method, release data and carry artifact are independently hashable.

What changed in v1.8.11

Added Minimodels Decision 1.0 (candy-cane), measured on the 27 Oct draw, unranked until the next ranking review.

Historical v1.3.0 board: weightings, per-task grid, topic radars and held-out diagnostics

Open this section to load the earlier public-only board and diagnostics.