dsh-eval-guard 0.2.3 → 0.3.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (221) hide show
  1. package/README.md +91 -13
  2. package/assets/commit0/instance.json +30 -0
  3. package/assets/commit0/workspace.zip +0 -0
  4. package/assets/featurebench/hidden/test_regression.py +690 -0
  5. package/assets/featurebench/instance.json +16 -0
  6. package/assets/featurebench/workspace.zip +0 -0
  7. package/assets/swe-bench-pro/hidden/test.patch +36 -0
  8. package/assets/swe-bench-pro/instance.json +10 -0
  9. package/assets/swe-bench-pro/workspace.zip +0 -0
  10. package/eval/README.md +18 -0
  11. package/eval/arena/arena/README.md +21 -0
  12. package/eval/arena/arena/bt.py +104 -0
  13. package/eval/arena/arena/clock.py +15 -0
  14. package/eval/arena/arena/core.py +265 -0
  15. package/eval/arena/arena/run.cmd +24 -0
  16. package/eval/arena/arena/run.sh +11 -0
  17. package/eval/arena/arena/schedule.py +79 -0
  18. package/eval/arena/arena/tournament.py +183 -0
  19. package/eval/arena/bomberman/bots/README.md +16 -0
  20. package/eval/arena/bomberman/bots/deepseek-v4.1-flash.cpp +447 -0
  21. package/eval/arena/bomberman/bots/doubao-seed-2.1.cpp +542 -0
  22. package/eval/arena/bomberman/bots/glm-5.3.cpp +406 -0
  23. package/eval/arena/bomberman/bots/gpt-6-astra.cpp +768 -0
  24. package/eval/arena/bomberman/bots/gpt-6-sol.cpp +444 -0
  25. package/eval/arena/bomberman/bots/kimi-k3.cpp +432 -0
  26. package/eval/arena/bomberman/bots/mimo-v2.6-pro.cpp +529 -0
  27. package/eval/arena/bomberman/bots/opus-5.5.cpp +548 -0
  28. package/eval/arena/bomberman/bots/qwen-3.8-max.cpp +836 -0
  29. package/eval/arena/bomberman/tournament/README.md +11 -0
  30. package/eval/arena/bomberman/tournament/bt.py +56 -0
  31. package/eval/arena/bomberman/tournament/evaluate.py +199 -0
  32. package/eval/arena/bomberman/tournament/run.py +79 -0
  33. package/eval/arena/bomberman/tournament/schedule-2to10.json.gz +0 -0
  34. package/eval/arena/bomberman/workspace/MANIFEST.json +13 -0
  35. package/eval/arena/bomberman/workspace/README.md +29 -0
  36. package/eval/arena/bomberman/workspace/RULES.md +100 -0
  37. package/eval/arena/bomberman/workspace/baseline.cpp +240 -0
  38. package/eval/arena/bomberman/workspace/clock.py +25 -0
  39. package/eval/arena/bomberman/workspace/engine/arena.py +198 -0
  40. package/eval/arena/bomberman/workspace/engine/build.py +35 -0
  41. package/eval/arena/bomberman/workspace/engine/engine.py +256 -0
  42. package/eval/arena/bomberman/workspace/engine/verify_replay.py +24 -0
  43. package/eval/arena/bomberman/workspace/run.cmd +28 -0
  44. package/eval/arena/bomberman/workspace/run.sh +16 -0
  45. package/eval/arena/bomberman/workspace/viewer/index.html +42 -0
  46. package/eval/arena/ctf/bots/README.md +18 -0
  47. package/eval/arena/ctf/bots/deepseek-v4.1-flash.cpp +601 -0
  48. package/eval/arena/ctf/bots/doubao-seed-2.1-turbo.cpp +320 -0
  49. package/eval/arena/ctf/bots/fable-5.1.cpp +634 -0
  50. package/eval/arena/ctf/bots/glm-5.3.cpp +689 -0
  51. package/eval/arena/ctf/bots/gpt-6-astra.cpp +657 -0
  52. package/eval/arena/ctf/bots/gpt-6.1-sol.cpp +725 -0
  53. package/eval/arena/ctf/bots/kimi-k3.cpp +667 -0
  54. package/eval/arena/ctf/bots/mimo-v2.6-pro.cpp +533 -0
  55. package/eval/arena/ctf/bots/opus-5.5.cpp +645 -0
  56. package/eval/arena/ctf/bots/qwen-3.8-max.cpp +726 -0
  57. package/eval/arena/ctf/bots/sonnet-5.5.cpp +569 -0
  58. package/eval/arena/ctf/tournament/README.md +15 -0
  59. package/eval/arena/ctf/tournament/formal-2to8.json +31 -0
  60. package/eval/arena/ctf/tournament/formal-9to15.json +31 -0
  61. package/eval/arena/ctf/workspace/README.md +41 -0
  62. package/eval/arena/ctf/workspace/RULES.md +119 -0
  63. package/eval/arena/ctf/workspace/arena.py +186 -0
  64. package/eval/arena/ctf/workspace/baseline.cpp +257 -0
  65. package/eval/arena/ctf/workspace/clock.py +15 -0
  66. package/eval/arena/ctf/workspace/core.py +265 -0
  67. package/eval/arena/ctf/workspace/engine.py +389 -0
  68. package/eval/arena/ctf/workspace/run.cmd +24 -0
  69. package/eval/arena/ctf/workspace/run.sh +11 -0
  70. package/eval/arena/ctf/workspace/viewer.html +71 -0
  71. package/eval/arena/land/bots/README.md +17 -0
  72. package/eval/arena/land/bots/deepseek-4.1f@workbuddy#1/bot.cpp +608 -0
  73. package/eval/arena/land/bots/glm-5.3@workbuddy#1/bot.cpp +521 -0
  74. package/eval/arena/land/bots/gpt-6-astra@codex#1/bot.cpp +373 -0
  75. package/eval/arena/land/bots/gpt-6-sol@codex#1/bot.cpp +341 -0
  76. package/eval/arena/land/bots/kimi-k3@workbuddy#1/bot.cpp +492 -0
  77. package/eval/arena/land/bots/mimo-2.6-pro@zcode#1/bot.cpp +652 -0
  78. package/eval/arena/land/bots/mystery-model@zcode#1/bot.cpp +466 -0
  79. package/eval/arena/land/bots/opus-5.5@claude-code#1/bot.cpp +630 -0
  80. package/eval/arena/land/bots/sonnet-5@claude-code#1/bot.cpp +619 -0
  81. package/eval/arena/land/tournament/make_replay_page.py +17 -0
  82. package/eval/arena/land/tournament/rate.py +290 -0
  83. package/eval/arena/land/tournament/round_robin.py +172 -0
  84. package/eval/arena/land/tournament/sparring.cpp +356 -0
  85. package/eval/arena/land/workspace/RULES.md +114 -0
  86. package/eval/arena/land/workspace/TASK.md +46 -0
  87. package/eval/arena/land/workspace/anchors/.build/greedy_f2e6d2dea5ca +0 -0
  88. package/eval/arena/land/workspace/anchors/greedy.cpp +194 -0
  89. package/eval/arena/land/workspace/anchors/random.cpp +80 -0
  90. package/eval/arena/land/workspace/clock.py +64 -0
  91. package/eval/arena/land/workspace/engine/build.py +33 -0
  92. package/eval/arena/land/workspace/engine/game.py +273 -0
  93. package/eval/arena/land/workspace/engine/referee.py +308 -0
  94. package/eval/arena/land/workspace/starter/.build/bot_be32b45b4699 +0 -0
  95. package/eval/arena/land/workspace/starter/bot.cpp +82 -0
  96. package/eval/arena/land/workspace/viewer/viewer.html +210 -0
  97. package/eval/arena/tank/bots/README.md +24 -0
  98. package/eval/arena/tank/bots/half1/deepseek-v4.1-flash.cpp +367 -0
  99. package/eval/arena/tank/bots/half1/doubao-seed-evolving.cpp +402 -0
  100. package/eval/arena/tank/bots/half1/gemini-4-argon.cpp +1132 -0
  101. package/eval/arena/tank/bots/half1/glm-5.3.cpp +551 -0
  102. package/eval/arena/tank/bots/half1/gpt-6-astra.cpp +233 -0
  103. package/eval/arena/tank/bots/half1/gpt-6.1-sol-ultrafast.cpp +350 -0
  104. package/eval/arena/tank/bots/half1/gpt-6.1-sol.cpp +78 -0
  105. package/eval/arena/tank/bots/half1/grok-4.7.cpp +1099 -0
  106. package/eval/arena/tank/bots/half1/haiku-5.5.cpp +210 -0
  107. package/eval/arena/tank/bots/half1/kimi-k3.cpp +353 -0
  108. package/eval/arena/tank/bots/half1/mimo-v2.6-pro.cpp +336 -0
  109. package/eval/arena/tank/bots/half1/muse-wonder-full.cpp +560 -0
  110. package/eval/arena/tank/bots/half1/opus-5.5.cpp +435 -0
  111. package/eval/arena/tank/bots/half1/qwen-3.8-max.cpp +537 -0
  112. package/eval/arena/tank/bots/half2/deepseek-v4.1-flash.cpp +425 -0
  113. package/eval/arena/tank/bots/half2/doubao-seed-evolving.cpp +308 -0
  114. package/eval/arena/tank/bots/half2/gemini-4-argon.cpp +1390 -0
  115. package/eval/arena/tank/bots/half2/glm-5.3.cpp +651 -0
  116. package/eval/arena/tank/bots/half2/gpt-6-astra.cpp +241 -0
  117. package/eval/arena/tank/bots/half2/gpt-6.1-sol-ultrafast.cpp +356 -0
  118. package/eval/arena/tank/bots/half2/gpt-6.1-sol.cpp +87 -0
  119. package/eval/arena/tank/bots/half2/grok-4.7.cpp +1205 -0
  120. package/eval/arena/tank/bots/half2/haiku-5.5.cpp +210 -0
  121. package/eval/arena/tank/bots/half2/kimi-k3.cpp +487 -0
  122. package/eval/arena/tank/bots/half2/mimo-v2.6-pro.cpp +336 -0
  123. package/eval/arena/tank/bots/half2/muse-wonder-full.cpp +560 -0
  124. package/eval/arena/tank/bots/half2/opus-5.5.cpp +476 -0
  125. package/eval/arena/tank/bots/half2/qwen-3.8-max.cpp +392 -0
  126. package/eval/arena/tank/tournament/README.md +24 -0
  127. package/eval/arena/tank/tournament/bt.py +218 -0
  128. package/eval/arena/tank/tournament/final.json +202 -0
  129. package/eval/arena/tank/tournament/halftime.json +146 -0
  130. package/eval/arena/tank/tournament/progress.json +911 -0
  131. package/eval/arena/tank/tournament/schedule.py +100 -0
  132. package/eval/arena/tank/tournament/sparring.cpp +928 -0
  133. package/eval/arena/tank/tournament/tournament.py +1426 -0
  134. package/eval/arena/tank/workspace/README.md +169 -0
  135. package/eval/arena/tank/workspace/RULES.md +443 -0
  136. package/eval/arena/tank/workspace/arena.py +862 -0
  137. package/eval/arena/tank/workspace/baseline.cpp +259 -0
  138. package/eval/arena/tank/workspace/clock.py +103 -0
  139. package/eval/arena/tank/workspace/core.py +712 -0
  140. package/eval/arena/tank/workspace/engine.py +573 -0
  141. package/eval/arena/tank/workspace/history/.lock +0 -0
  142. package/eval/arena/tank/workspace/history/20261017-193610-08715287f91f-bot.cpp +109 -0
  143. package/eval/arena/tank/workspace/history/20261017-193619-f00b2d5635fa-gpt-6-astra.cpp +241 -0
  144. package/eval/arena/tank/workspace/history/20261017-193643-63319e0a2b1a-gpt-6_1-sol-ultrafast.cpp +356 -0
  145. package/eval/arena/tank/workspace/history/20261017-193647-41c1341d0a8a-gemini-4-argon.cpp +1390 -0
  146. package/eval/arena/tank/workspace/history/20261017-193652-e34c0f45539a-gpt-6_1-sol.cpp +87 -0
  147. package/eval/arena/tank/workspace/history/20261017-193654-c026348ce96e-opus-5_5.cpp +476 -0
  148. package/eval/arena/tank/workspace/history/20261017-193655-4a21366e94d1-muse-wonder-full.cpp +560 -0
  149. package/eval/arena/tank/workspace/history/20261017-193657-dbc4e9969c52-mimo-v2_6-pro.cpp +336 -0
  150. package/eval/arena/tank/workspace/history/20261017-193703-723cd036eb31-grok-4_7.cpp +1205 -0
  151. package/eval/arena/tank/workspace/history/20261017-193707-1ebbb227be68-deepseek-v4_1-flash.cpp +425 -0
  152. package/eval/arena/tank/workspace/history/20261017-193709-aeb0e7202ad5-doubao-seed-evolving.cpp +308 -0
  153. package/eval/arena/tank/workspace/history/20261017-193710-b21ebeb9bc76-kimi-k3.cpp +487 -0
  154. package/eval/arena/tank/workspace/history/20261017-193712-54cd86531905-haiku-5_5.cpp +210 -0
  155. package/eval/arena/tank/workspace/history/20261017-193714-efdb8e35c0bf-qwen-3_8-max.cpp +392 -0
  156. package/eval/arena/tank/workspace/history/20261017-193715-681423c23fdb-glm-5_3.cpp +651 -0
  157. package/eval/arena/tank/workspace/history/20261017-193717-1a8b3c75181d-sparring.cpp +928 -0
  158. package/eval/arena/tank/workspace/history/20261017-193720-1e0803a541e8-baseline.cpp +259 -0
  159. package/eval/arena/tank/workspace/history/log.jsonl +103 -0
  160. package/eval/arena/tank/workspace/judge/cycles.hpp +53 -0
  161. package/eval/arena/tank/workspace/judge/fast.cpp +337 -0
  162. package/eval/arena/tank/workspace/judge/referee.cpp +125 -0
  163. package/eval/arena/tank/workspace/judge/speed.cpp +99 -0
  164. package/eval/arena/tank/workspace/run.cmd +31 -0
  165. package/eval/arena/tank/workspace/run.sh +23 -0
  166. package/eval/arena/tank/workspace/tank.hpp +1762 -0
  167. package/eval/arena/tank/workspace/tank_main.cpp +215 -0
  168. package/eval/arena/tank/workspace/viewer.html +1150 -0
  169. package/eval/arena/warehouse/workspace/SPEC.md +130 -0
  170. package/eval/arena/warehouse/workspace/TASK.md +71 -0
  171. package/eval/arena/warehouse/workspace/baseline/common.h +88 -0
  172. package/eval/arena/warehouse/workspace/baseline/main.cpp +27 -0
  173. package/eval/arena/warehouse/workspace/baseline/old_stuff.cpp +87 -0
  174. package/eval/arena/warehouse/workspace/baseline/sim.cpp +427 -0
  175. package/eval/arena/warehouse/workspace/baseline/util.cpp +220 -0
  176. package/eval/arena/warehouse/workspace/clock.py +42 -0
  177. package/eval/arena/warehouse/workspace/src/common.h +88 -0
  178. package/eval/arena/warehouse/workspace/src/main.cpp +27 -0
  179. package/eval/arena/warehouse/workspace/src/old_stuff.cpp +87 -0
  180. package/eval/arena/warehouse/workspace/src/sim.cpp +427 -0
  181. package/eval/arena/warehouse/workspace/src/util.cpp +220 -0
  182. package/eval/arena/warehouse/workspace/tools/bench.py +119 -0
  183. package/eval/arena/warehouse/workspace/tools/check_input.py +161 -0
  184. package/eval/arena/warehouse/workspace/tools/gen.py +132 -0
  185. package/eval/arena/warehouse/workspace/tools/measure.py +107 -0
  186. package/eval/arena/warehouse/workspace/workloads/public/pub_crowded.in +1420 -0
  187. package/eval/arena/warehouse/workspace/workloads/public/pub_crowded.out +9162 -0
  188. package/eval/arena/warehouse/workspace/workloads/public/pub_dynamic.in +1465 -0
  189. package/eval/arena/warehouse/workspace/workloads/public/pub_dynamic.out +3439 -0
  190. package/eval/arena/warehouse/workspace/workloads/public/pub_flood.in +2757 -0
  191. package/eval/arena/warehouse/workspace/workloads/public/pub_flood.out +3126 -0
  192. package/eval/arena/warehouse/workspace/workloads/public/pub_small.in +731 -0
  193. package/eval/arena/warehouse/workspace/workloads/public/pub_small.out +2578 -0
  194. package/eval/arena/warehouse/workspace/workloads/public/pub_wide.in +528 -0
  195. package/eval/arena/warehouse/workspace/workloads/public/pub_wide.out +2026 -0
  196. package/eval/bomberman.py +20 -0
  197. package/eval/commit0.py +21 -0
  198. package/eval/common.py +61 -0
  199. package/eval/ctf.py +20 -0
  200. package/eval/featurebench.py +25 -0
  201. package/eval/land.py +20 -0
  202. package/eval/legacy/bomberman-prices.json +4 -0
  203. package/eval/legacy/bomberman.py +161 -0
  204. package/eval/legacy/commit0.py +91 -0
  205. package/eval/legacy/ctf-prices.json +4 -0
  206. package/eval/legacy/ctf.py +178 -0
  207. package/eval/legacy/featurebench.py +103 -0
  208. package/eval/legacy/land-prices.json +4 -0
  209. package/eval/legacy/land.py +177 -0
  210. package/eval/legacy/swe_bench_pro.py +116 -0
  211. package/eval/legacy/tank-prices.json +3 -0
  212. package/eval/legacy/tank.py +408 -0
  213. package/eval/legacy/warehouse-prices.json +4 -0
  214. package/eval/legacy/warehouse.py +365 -0
  215. package/eval/swe-bench-pro.py +25 -0
  216. package/eval/tank.py +20 -0
  217. package/eval/warehouse.py +20 -0
  218. package/lib/client/index.js +376 -63
  219. package/lib/client/index.js.map +1 -1
  220. package/lib/index.js +684 -27
  221. package/package.json +2 -1
@@ -0,0 +1,911 @@
1
+ {
2
+ "title": "进步赛(下半场版本 对 全员上半场版本 + 基准)",
3
+ "kit": "../workspace",
4
+ "build_dir": "../workspace/build",
5
+ "out": "run-progress",
6
+ "players": {
7
+ "Astra@2": "../bots/half2/gpt-6-astra.cpp",
8
+ "Opus@2": "../bots/half2/opus-5.5.cpp",
9
+ "Haiku@2": "../bots/half2/haiku-5.5.cpp",
10
+ "Sol@2": "../bots/half2/gpt-6.1-sol.cpp",
11
+ "GLM@2": "../bots/half2/glm-5.3.cpp",
12
+ "MiMo@2": "../bots/half2/mimo-v2.6-pro.cpp",
13
+ "Doubao@2": "../bots/half2/doubao-seed-evolving.cpp",
14
+ "DeepSeek@2": "../bots/half2/deepseek-v4.1-flash.cpp",
15
+ "Muse@2": "../bots/half2/muse-wonder-full.cpp",
16
+ "Grok@2": "../bots/half2/grok-4.7.cpp",
17
+ "Qwen@2": "../bots/half2/qwen-3.8-max.cpp",
18
+ "Kimi@2": "../bots/half2/kimi-k3.cpp",
19
+ "SolUF@2": "../bots/half2/gpt-6.1-sol-ultrafast.cpp",
20
+ "Gemini@2": "../bots/half2/gemini-4-argon.cpp",
21
+ "Astra@1": "../bots/half1/gpt-6-astra.cpp",
22
+ "Opus@1": "../bots/half1/opus-5.5.cpp",
23
+ "Haiku@1": "../bots/half1/haiku-5.5.cpp",
24
+ "Sol@1": "../bots/half1/gpt-6.1-sol.cpp",
25
+ "GLM@1": "../bots/half1/glm-5.3.cpp",
26
+ "MiMo@1": "../bots/half1/mimo-v2.6-pro.cpp",
27
+ "Doubao@1": "../bots/half1/doubao-seed-evolving.cpp",
28
+ "DeepSeek@1": "../bots/half1/deepseek-v4.1-flash.cpp",
29
+ "Muse@1": "../bots/half1/muse-wonder-full.cpp",
30
+ "Grok@1": "../bots/half1/grok-4.7.cpp",
31
+ "Qwen@1": "../bots/half1/qwen-3.8-max.cpp",
32
+ "Kimi@1": "../bots/half1/kimi-k3.cpp",
33
+ "SolUF@1": "../bots/half1/gpt-6.1-sol-ultrafast.cpp",
34
+ "Gemini@1": "../bots/half1/gemini-4-argon.cpp",
35
+ "baseline": "../workspace/baseline.cpp"
36
+ },
37
+ "maps_per_pair": 8,
38
+ "workers": 0,
39
+ "pin": {
40
+ "mode": "big"
41
+ },
42
+ "speed": {
43
+ "enabled": true,
44
+ "tolerance": 0.15,
45
+ "repeat": 3
46
+ },
47
+ "boot": 1000,
48
+ "labels": {
49
+ "baseline": "基准",
50
+ "sparring": "出题方参照"
51
+ },
52
+ "display": {
53
+ "Astra": "GPT-6 Astra",
54
+ "Opus": "Claude Opus 5.5",
55
+ "Haiku": "Claude Haiku 5.5",
56
+ "Sol": "GPT-6.1 Sol",
57
+ "GLM": "智谱 GLM 5.3",
58
+ "MiMo": "小米 MiMo V2.6 Pro",
59
+ "Doubao": "Doubao-Seed-Evolving",
60
+ "DeepSeek": "DeepSeek V4.1 Flash",
61
+ "Muse": "Muse Wonder Full",
62
+ "Grok": "Grok 4.7",
63
+ "Qwen": "通义千问 Qwen 3.8 Max",
64
+ "Kimi": "Kimi K3",
65
+ "SolUF": "GPT-6.1 Sol Ultrafast(Codex CLI)",
66
+ "Gemini": "Gemini 4 Argon Contributor(OpenCode Go)"
67
+ },
68
+ "pairs": [
69
+ [
70
+ "Astra@2",
71
+ "Astra@1"
72
+ ],
73
+ [
74
+ "Astra@2",
75
+ "Opus@1"
76
+ ],
77
+ [
78
+ "Astra@2",
79
+ "Haiku@1"
80
+ ],
81
+ [
82
+ "Astra@2",
83
+ "Sol@1"
84
+ ],
85
+ [
86
+ "Astra@2",
87
+ "GLM@1"
88
+ ],
89
+ [
90
+ "Astra@2",
91
+ "MiMo@1"
92
+ ],
93
+ [
94
+ "Astra@2",
95
+ "Doubao@1"
96
+ ],
97
+ [
98
+ "Astra@2",
99
+ "DeepSeek@1"
100
+ ],
101
+ [
102
+ "Astra@2",
103
+ "Muse@1"
104
+ ],
105
+ [
106
+ "Astra@2",
107
+ "Grok@1"
108
+ ],
109
+ [
110
+ "Astra@2",
111
+ "Qwen@1"
112
+ ],
113
+ [
114
+ "Astra@2",
115
+ "Kimi@1"
116
+ ],
117
+ [
118
+ "Astra@2",
119
+ "SolUF@1"
120
+ ],
121
+ [
122
+ "Astra@2",
123
+ "Gemini@1"
124
+ ],
125
+ [
126
+ "Astra@2",
127
+ "baseline"
128
+ ],
129
+ [
130
+ "Opus@2",
131
+ "Astra@1"
132
+ ],
133
+ [
134
+ "Opus@2",
135
+ "Opus@1"
136
+ ],
137
+ [
138
+ "Opus@2",
139
+ "Haiku@1"
140
+ ],
141
+ [
142
+ "Opus@2",
143
+ "Sol@1"
144
+ ],
145
+ [
146
+ "Opus@2",
147
+ "GLM@1"
148
+ ],
149
+ [
150
+ "Opus@2",
151
+ "MiMo@1"
152
+ ],
153
+ [
154
+ "Opus@2",
155
+ "Doubao@1"
156
+ ],
157
+ [
158
+ "Opus@2",
159
+ "DeepSeek@1"
160
+ ],
161
+ [
162
+ "Opus@2",
163
+ "Muse@1"
164
+ ],
165
+ [
166
+ "Opus@2",
167
+ "Grok@1"
168
+ ],
169
+ [
170
+ "Opus@2",
171
+ "Qwen@1"
172
+ ],
173
+ [
174
+ "Opus@2",
175
+ "Kimi@1"
176
+ ],
177
+ [
178
+ "Opus@2",
179
+ "SolUF@1"
180
+ ],
181
+ [
182
+ "Opus@2",
183
+ "Gemini@1"
184
+ ],
185
+ [
186
+ "Opus@2",
187
+ "baseline"
188
+ ],
189
+ [
190
+ "Haiku@2",
191
+ "Astra@1"
192
+ ],
193
+ [
194
+ "Haiku@2",
195
+ "Opus@1"
196
+ ],
197
+ [
198
+ "Haiku@2",
199
+ "Haiku@1"
200
+ ],
201
+ [
202
+ "Haiku@2",
203
+ "Sol@1"
204
+ ],
205
+ [
206
+ "Haiku@2",
207
+ "GLM@1"
208
+ ],
209
+ [
210
+ "Haiku@2",
211
+ "MiMo@1"
212
+ ],
213
+ [
214
+ "Haiku@2",
215
+ "Doubao@1"
216
+ ],
217
+ [
218
+ "Haiku@2",
219
+ "DeepSeek@1"
220
+ ],
221
+ [
222
+ "Haiku@2",
223
+ "Muse@1"
224
+ ],
225
+ [
226
+ "Haiku@2",
227
+ "Grok@1"
228
+ ],
229
+ [
230
+ "Haiku@2",
231
+ "Qwen@1"
232
+ ],
233
+ [
234
+ "Haiku@2",
235
+ "Kimi@1"
236
+ ],
237
+ [
238
+ "Haiku@2",
239
+ "SolUF@1"
240
+ ],
241
+ [
242
+ "Haiku@2",
243
+ "Gemini@1"
244
+ ],
245
+ [
246
+ "Haiku@2",
247
+ "baseline"
248
+ ],
249
+ [
250
+ "Sol@2",
251
+ "Astra@1"
252
+ ],
253
+ [
254
+ "Sol@2",
255
+ "Opus@1"
256
+ ],
257
+ [
258
+ "Sol@2",
259
+ "Haiku@1"
260
+ ],
261
+ [
262
+ "Sol@2",
263
+ "Sol@1"
264
+ ],
265
+ [
266
+ "Sol@2",
267
+ "GLM@1"
268
+ ],
269
+ [
270
+ "Sol@2",
271
+ "MiMo@1"
272
+ ],
273
+ [
274
+ "Sol@2",
275
+ "Doubao@1"
276
+ ],
277
+ [
278
+ "Sol@2",
279
+ "DeepSeek@1"
280
+ ],
281
+ [
282
+ "Sol@2",
283
+ "Muse@1"
284
+ ],
285
+ [
286
+ "Sol@2",
287
+ "Grok@1"
288
+ ],
289
+ [
290
+ "Sol@2",
291
+ "Qwen@1"
292
+ ],
293
+ [
294
+ "Sol@2",
295
+ "Kimi@1"
296
+ ],
297
+ [
298
+ "Sol@2",
299
+ "SolUF@1"
300
+ ],
301
+ [
302
+ "Sol@2",
303
+ "Gemini@1"
304
+ ],
305
+ [
306
+ "Sol@2",
307
+ "baseline"
308
+ ],
309
+ [
310
+ "GLM@2",
311
+ "Astra@1"
312
+ ],
313
+ [
314
+ "GLM@2",
315
+ "Opus@1"
316
+ ],
317
+ [
318
+ "GLM@2",
319
+ "Haiku@1"
320
+ ],
321
+ [
322
+ "GLM@2",
323
+ "Sol@1"
324
+ ],
325
+ [
326
+ "GLM@2",
327
+ "GLM@1"
328
+ ],
329
+ [
330
+ "GLM@2",
331
+ "MiMo@1"
332
+ ],
333
+ [
334
+ "GLM@2",
335
+ "Doubao@1"
336
+ ],
337
+ [
338
+ "GLM@2",
339
+ "DeepSeek@1"
340
+ ],
341
+ [
342
+ "GLM@2",
343
+ "Muse@1"
344
+ ],
345
+ [
346
+ "GLM@2",
347
+ "Grok@1"
348
+ ],
349
+ [
350
+ "GLM@2",
351
+ "Qwen@1"
352
+ ],
353
+ [
354
+ "GLM@2",
355
+ "Kimi@1"
356
+ ],
357
+ [
358
+ "GLM@2",
359
+ "SolUF@1"
360
+ ],
361
+ [
362
+ "GLM@2",
363
+ "Gemini@1"
364
+ ],
365
+ [
366
+ "GLM@2",
367
+ "baseline"
368
+ ],
369
+ [
370
+ "MiMo@2",
371
+ "Astra@1"
372
+ ],
373
+ [
374
+ "MiMo@2",
375
+ "Opus@1"
376
+ ],
377
+ [
378
+ "MiMo@2",
379
+ "Haiku@1"
380
+ ],
381
+ [
382
+ "MiMo@2",
383
+ "Sol@1"
384
+ ],
385
+ [
386
+ "MiMo@2",
387
+ "GLM@1"
388
+ ],
389
+ [
390
+ "MiMo@2",
391
+ "MiMo@1"
392
+ ],
393
+ [
394
+ "MiMo@2",
395
+ "Doubao@1"
396
+ ],
397
+ [
398
+ "MiMo@2",
399
+ "DeepSeek@1"
400
+ ],
401
+ [
402
+ "MiMo@2",
403
+ "Muse@1"
404
+ ],
405
+ [
406
+ "MiMo@2",
407
+ "Grok@1"
408
+ ],
409
+ [
410
+ "MiMo@2",
411
+ "Qwen@1"
412
+ ],
413
+ [
414
+ "MiMo@2",
415
+ "Kimi@1"
416
+ ],
417
+ [
418
+ "MiMo@2",
419
+ "SolUF@1"
420
+ ],
421
+ [
422
+ "MiMo@2",
423
+ "Gemini@1"
424
+ ],
425
+ [
426
+ "MiMo@2",
427
+ "baseline"
428
+ ],
429
+ [
430
+ "Doubao@2",
431
+ "Astra@1"
432
+ ],
433
+ [
434
+ "Doubao@2",
435
+ "Opus@1"
436
+ ],
437
+ [
438
+ "Doubao@2",
439
+ "Haiku@1"
440
+ ],
441
+ [
442
+ "Doubao@2",
443
+ "Sol@1"
444
+ ],
445
+ [
446
+ "Doubao@2",
447
+ "GLM@1"
448
+ ],
449
+ [
450
+ "Doubao@2",
451
+ "MiMo@1"
452
+ ],
453
+ [
454
+ "Doubao@2",
455
+ "Doubao@1"
456
+ ],
457
+ [
458
+ "Doubao@2",
459
+ "DeepSeek@1"
460
+ ],
461
+ [
462
+ "Doubao@2",
463
+ "Muse@1"
464
+ ],
465
+ [
466
+ "Doubao@2",
467
+ "Grok@1"
468
+ ],
469
+ [
470
+ "Doubao@2",
471
+ "Qwen@1"
472
+ ],
473
+ [
474
+ "Doubao@2",
475
+ "Kimi@1"
476
+ ],
477
+ [
478
+ "Doubao@2",
479
+ "SolUF@1"
480
+ ],
481
+ [
482
+ "Doubao@2",
483
+ "Gemini@1"
484
+ ],
485
+ [
486
+ "Doubao@2",
487
+ "baseline"
488
+ ],
489
+ [
490
+ "DeepSeek@2",
491
+ "Astra@1"
492
+ ],
493
+ [
494
+ "DeepSeek@2",
495
+ "Opus@1"
496
+ ],
497
+ [
498
+ "DeepSeek@2",
499
+ "Haiku@1"
500
+ ],
501
+ [
502
+ "DeepSeek@2",
503
+ "Sol@1"
504
+ ],
505
+ [
506
+ "DeepSeek@2",
507
+ "GLM@1"
508
+ ],
509
+ [
510
+ "DeepSeek@2",
511
+ "MiMo@1"
512
+ ],
513
+ [
514
+ "DeepSeek@2",
515
+ "Doubao@1"
516
+ ],
517
+ [
518
+ "DeepSeek@2",
519
+ "DeepSeek@1"
520
+ ],
521
+ [
522
+ "DeepSeek@2",
523
+ "Muse@1"
524
+ ],
525
+ [
526
+ "DeepSeek@2",
527
+ "Grok@1"
528
+ ],
529
+ [
530
+ "DeepSeek@2",
531
+ "Qwen@1"
532
+ ],
533
+ [
534
+ "DeepSeek@2",
535
+ "Kimi@1"
536
+ ],
537
+ [
538
+ "DeepSeek@2",
539
+ "SolUF@1"
540
+ ],
541
+ [
542
+ "DeepSeek@2",
543
+ "Gemini@1"
544
+ ],
545
+ [
546
+ "DeepSeek@2",
547
+ "baseline"
548
+ ],
549
+ [
550
+ "Muse@2",
551
+ "Astra@1"
552
+ ],
553
+ [
554
+ "Muse@2",
555
+ "Opus@1"
556
+ ],
557
+ [
558
+ "Muse@2",
559
+ "Haiku@1"
560
+ ],
561
+ [
562
+ "Muse@2",
563
+ "Sol@1"
564
+ ],
565
+ [
566
+ "Muse@2",
567
+ "GLM@1"
568
+ ],
569
+ [
570
+ "Muse@2",
571
+ "MiMo@1"
572
+ ],
573
+ [
574
+ "Muse@2",
575
+ "Doubao@1"
576
+ ],
577
+ [
578
+ "Muse@2",
579
+ "DeepSeek@1"
580
+ ],
581
+ [
582
+ "Muse@2",
583
+ "Muse@1"
584
+ ],
585
+ [
586
+ "Muse@2",
587
+ "Grok@1"
588
+ ],
589
+ [
590
+ "Muse@2",
591
+ "Qwen@1"
592
+ ],
593
+ [
594
+ "Muse@2",
595
+ "Kimi@1"
596
+ ],
597
+ [
598
+ "Muse@2",
599
+ "SolUF@1"
600
+ ],
601
+ [
602
+ "Muse@2",
603
+ "Gemini@1"
604
+ ],
605
+ [
606
+ "Muse@2",
607
+ "baseline"
608
+ ],
609
+ [
610
+ "Grok@2",
611
+ "Astra@1"
612
+ ],
613
+ [
614
+ "Grok@2",
615
+ "Opus@1"
616
+ ],
617
+ [
618
+ "Grok@2",
619
+ "Haiku@1"
620
+ ],
621
+ [
622
+ "Grok@2",
623
+ "Sol@1"
624
+ ],
625
+ [
626
+ "Grok@2",
627
+ "GLM@1"
628
+ ],
629
+ [
630
+ "Grok@2",
631
+ "MiMo@1"
632
+ ],
633
+ [
634
+ "Grok@2",
635
+ "Doubao@1"
636
+ ],
637
+ [
638
+ "Grok@2",
639
+ "DeepSeek@1"
640
+ ],
641
+ [
642
+ "Grok@2",
643
+ "Muse@1"
644
+ ],
645
+ [
646
+ "Grok@2",
647
+ "Grok@1"
648
+ ],
649
+ [
650
+ "Grok@2",
651
+ "Qwen@1"
652
+ ],
653
+ [
654
+ "Grok@2",
655
+ "Kimi@1"
656
+ ],
657
+ [
658
+ "Grok@2",
659
+ "SolUF@1"
660
+ ],
661
+ [
662
+ "Grok@2",
663
+ "Gemini@1"
664
+ ],
665
+ [
666
+ "Grok@2",
667
+ "baseline"
668
+ ],
669
+ [
670
+ "Qwen@2",
671
+ "Astra@1"
672
+ ],
673
+ [
674
+ "Qwen@2",
675
+ "Opus@1"
676
+ ],
677
+ [
678
+ "Qwen@2",
679
+ "Haiku@1"
680
+ ],
681
+ [
682
+ "Qwen@2",
683
+ "Sol@1"
684
+ ],
685
+ [
686
+ "Qwen@2",
687
+ "GLM@1"
688
+ ],
689
+ [
690
+ "Qwen@2",
691
+ "MiMo@1"
692
+ ],
693
+ [
694
+ "Qwen@2",
695
+ "Doubao@1"
696
+ ],
697
+ [
698
+ "Qwen@2",
699
+ "DeepSeek@1"
700
+ ],
701
+ [
702
+ "Qwen@2",
703
+ "Muse@1"
704
+ ],
705
+ [
706
+ "Qwen@2",
707
+ "Grok@1"
708
+ ],
709
+ [
710
+ "Qwen@2",
711
+ "Qwen@1"
712
+ ],
713
+ [
714
+ "Qwen@2",
715
+ "Kimi@1"
716
+ ],
717
+ [
718
+ "Qwen@2",
719
+ "SolUF@1"
720
+ ],
721
+ [
722
+ "Qwen@2",
723
+ "Gemini@1"
724
+ ],
725
+ [
726
+ "Qwen@2",
727
+ "baseline"
728
+ ],
729
+ [
730
+ "Kimi@2",
731
+ "Astra@1"
732
+ ],
733
+ [
734
+ "Kimi@2",
735
+ "Opus@1"
736
+ ],
737
+ [
738
+ "Kimi@2",
739
+ "Haiku@1"
740
+ ],
741
+ [
742
+ "Kimi@2",
743
+ "Sol@1"
744
+ ],
745
+ [
746
+ "Kimi@2",
747
+ "GLM@1"
748
+ ],
749
+ [
750
+ "Kimi@2",
751
+ "MiMo@1"
752
+ ],
753
+ [
754
+ "Kimi@2",
755
+ "Doubao@1"
756
+ ],
757
+ [
758
+ "Kimi@2",
759
+ "DeepSeek@1"
760
+ ],
761
+ [
762
+ "Kimi@2",
763
+ "Muse@1"
764
+ ],
765
+ [
766
+ "Kimi@2",
767
+ "Grok@1"
768
+ ],
769
+ [
770
+ "Kimi@2",
771
+ "Qwen@1"
772
+ ],
773
+ [
774
+ "Kimi@2",
775
+ "Kimi@1"
776
+ ],
777
+ [
778
+ "Kimi@2",
779
+ "SolUF@1"
780
+ ],
781
+ [
782
+ "Kimi@2",
783
+ "Gemini@1"
784
+ ],
785
+ [
786
+ "Kimi@2",
787
+ "baseline"
788
+ ],
789
+ [
790
+ "SolUF@2",
791
+ "Astra@1"
792
+ ],
793
+ [
794
+ "SolUF@2",
795
+ "Opus@1"
796
+ ],
797
+ [
798
+ "SolUF@2",
799
+ "Haiku@1"
800
+ ],
801
+ [
802
+ "SolUF@2",
803
+ "Sol@1"
804
+ ],
805
+ [
806
+ "SolUF@2",
807
+ "GLM@1"
808
+ ],
809
+ [
810
+ "SolUF@2",
811
+ "MiMo@1"
812
+ ],
813
+ [
814
+ "SolUF@2",
815
+ "Doubao@1"
816
+ ],
817
+ [
818
+ "SolUF@2",
819
+ "DeepSeek@1"
820
+ ],
821
+ [
822
+ "SolUF@2",
823
+ "Muse@1"
824
+ ],
825
+ [
826
+ "SolUF@2",
827
+ "Grok@1"
828
+ ],
829
+ [
830
+ "SolUF@2",
831
+ "Qwen@1"
832
+ ],
833
+ [
834
+ "SolUF@2",
835
+ "Kimi@1"
836
+ ],
837
+ [
838
+ "SolUF@2",
839
+ "SolUF@1"
840
+ ],
841
+ [
842
+ "SolUF@2",
843
+ "Gemini@1"
844
+ ],
845
+ [
846
+ "SolUF@2",
847
+ "baseline"
848
+ ],
849
+ [
850
+ "Gemini@2",
851
+ "Astra@1"
852
+ ],
853
+ [
854
+ "Gemini@2",
855
+ "Opus@1"
856
+ ],
857
+ [
858
+ "Gemini@2",
859
+ "Haiku@1"
860
+ ],
861
+ [
862
+ "Gemini@2",
863
+ "Sol@1"
864
+ ],
865
+ [
866
+ "Gemini@2",
867
+ "GLM@1"
868
+ ],
869
+ [
870
+ "Gemini@2",
871
+ "MiMo@1"
872
+ ],
873
+ [
874
+ "Gemini@2",
875
+ "Doubao@1"
876
+ ],
877
+ [
878
+ "Gemini@2",
879
+ "DeepSeek@1"
880
+ ],
881
+ [
882
+ "Gemini@2",
883
+ "Muse@1"
884
+ ],
885
+ [
886
+ "Gemini@2",
887
+ "Grok@1"
888
+ ],
889
+ [
890
+ "Gemini@2",
891
+ "Qwen@1"
892
+ ],
893
+ [
894
+ "Gemini@2",
895
+ "Kimi@1"
896
+ ],
897
+ [
898
+ "Gemini@2",
899
+ "SolUF@1"
900
+ ],
901
+ [
902
+ "Gemini@2",
903
+ "Gemini@1"
904
+ ],
905
+ [
906
+ "Gemini@2",
907
+ "baseline"
908
+ ]
909
+ ],
910
+ "seed": 95350690457500
911
+ }