@zhivex-ai/harness 1.0.0 → 1.1.0-rc.2

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 (181) hide show
  1. package/CHANGELOG.md +97 -1
  2. package/README.md +52 -306
  3. package/SECURITY.md +1 -1
  4. package/SUPPORT.md +7 -3
  5. package/benchmarks/README.md +3 -3
  6. package/contracts/client-protocol-v1.schema.json +341 -0
  7. package/contracts/public-api.json +100 -1
  8. package/contracts/stable-api-signatures.json +14 -14
  9. package/dist/approval-diff.d.ts +45 -0
  10. package/dist/approval-history.d.ts +38 -0
  11. package/dist/approval-preview.d.ts +26 -0
  12. package/dist/cli-credentials.d.ts +38 -0
  13. package/dist/cli-help.d.ts +3 -0
  14. package/dist/cli-options.d.ts +1 -1
  15. package/dist/cli-stream.d.ts +3 -3
  16. package/dist/cli.d.ts +15 -69
  17. package/dist/cli.js +13796 -10032
  18. package/dist/client-contract.d.ts +208 -0
  19. package/dist/console-commands.d.ts +6 -0
  20. package/dist/console-context.d.ts +5 -1
  21. package/dist/console-input.d.ts +44 -4
  22. package/dist/console-model-catalog.d.ts +25 -0
  23. package/dist/console-navigation.d.ts +27 -0
  24. package/dist/console-presentation.d.ts +18 -0
  25. package/dist/console-welcome.d.ts +13 -0
  26. package/dist/execution-environment.d.ts +4 -0
  27. package/dist/harness.d.ts +3 -0
  28. package/dist/hostile-repository-demo.js +1913 -798
  29. package/dist/index.d.ts +7 -0
  30. package/dist/index.js +7908 -6330
  31. package/dist/local-service.d.ts +34 -0
  32. package/dist/operations.d.ts +23 -1
  33. package/dist/orchestration.d.ts +3 -0
  34. package/dist/persistence-secrets.d.ts +2 -0
  35. package/dist/recorded-workspace.d.ts +4 -0
  36. package/dist/run-document.d.ts +91 -0
  37. package/dist/runtime-policy.d.ts +4 -4
  38. package/dist/service-cli.d.ts +2 -0
  39. package/dist/service-cli.js +10858 -0
  40. package/dist/service-client-cli.d.ts +2 -0
  41. package/dist/service-events.d.ts +50 -0
  42. package/dist/sessions.d.ts +3 -0
  43. package/dist/sqlite-access.d.ts +26 -0
  44. package/dist/sqlite-database.d.ts +2 -0
  45. package/dist/state-backup.d.ts +8 -0
  46. package/dist/task-memory.d.ts +2 -2
  47. package/dist/terminal-continuation.d.ts +3 -0
  48. package/dist/terminal-markdown.d.ts +1 -0
  49. package/dist/terminal-ui.d.ts +5 -0
  50. package/dist/usage-ledger.d.ts +84 -0
  51. package/dist/verify-historical-migrations.js +5699 -5191
  52. package/dist/workspace.d.ts +22 -0
  53. package/dist/zhx.js +13796 -10032
  54. package/docs/CLI.md +136 -30
  55. package/docs/CLIENT_PROTOCOL.md +212 -0
  56. package/docs/CREDENTIALS.md +59 -0
  57. package/docs/DURABLE_OPERATIONS.md +2 -2
  58. package/docs/FIRST_USE.md +82 -0
  59. package/docs/FIRST_USE_EXAMPLE.md +40 -0
  60. package/docs/HOSTILE_REPOSITORY_DEMO.md +1 -1
  61. package/docs/LOCAL_SERVICE.md +142 -0
  62. package/docs/README.md +24 -23
  63. package/docs/THREAT_MODEL.md +1 -1
  64. package/docs/TIME_TO_SAFE_FIX.md +2 -2
  65. package/docs/USAGE.md +181 -0
  66. package/evaluations/daily-cli-cohort.json +33 -0
  67. package/evaluations/representative-assembly-matrix.json +19 -1
  68. package/examples/README.md +26 -0
  69. package/examples/first-use/greeting.test.ts +3 -0
  70. package/examples/first-use/greeting.ts +1 -0
  71. package/examples/first-use/package.json +7 -0
  72. package/package.json +34 -9
  73. package/ROADMAP.md +0 -414
  74. package/benchmarks/baselines/context-navigation-2026-09-09.json +0 -94
  75. package/benchmarks/baselines/efficiency-2026-09-08.json +0 -233
  76. package/benchmarks/baselines/external-final-2026-09-08.json +0 -588
  77. package/benchmarks/baselines/external-intermediate-2026-09-08.json +0 -116
  78. package/benchmarks/baselines/external-matched-openai-2026-09-08.json +0 -1922
  79. package/benchmarks/baselines/external-matched-qwen-2026-09-08.json +0 -1706
  80. package/benchmarks/baselines/external-pilot-2026-09-08.json +0 -439
  81. package/benchmarks/baselines/external-qwen-flash-2026-09-08.json +0 -1570
  82. package/benchmarks/baselines/external-qwen-remediation-2026-09-09.json +0 -1509
  83. package/benchmarks/baselines/external-qwen-sdk-114-progress-2026-09-09.json +0 -2509
  84. package/benchmarks/baselines/external-sdk-upgrade-2026-09-08.json +0 -564
  85. package/benchmarks/baselines/external-verifier-recovery-2026-09-09.json +0 -794
  86. package/benchmarks/baselines/full-harness-review-2026-09-10.json +0 -134
  87. package/benchmarks/baselines/harness-audit-2026-09-09.json +0 -136
  88. package/benchmarks/baselines/harness-remediation-2026-09-09.json +0 -160
  89. package/benchmarks/baselines/oci-remediation-2026-09-10.json +0 -535
  90. package/benchmarks/baselines/rc14-holdout-2026-09-10.json +0 -854
  91. package/benchmarks/baselines/repair-architecture-audit-2026-09-09.json +0 -75
  92. package/benchmarks/baselines/time-to-safe-fix-live-expanded-2026-08-21.json +0 -186
  93. package/benchmarks/baselines/time-to-safe-fix-live-smoke-2026-08-21.json +0 -186
  94. package/docs/GA_READINESS.md +0 -29
  95. package/docs/GEMINI_1_0_DECISION.md +0 -9
  96. package/docs/LIVE_CERTIFICATION.md +0 -107
  97. package/docs/RC13_SECURITY_REVIEW.md +0 -28
  98. package/docs/RC14_DELEGATED_SECURITY_DECISION.md +0 -80
  99. package/docs/RC14_SECURITY_REVIEW.md +0 -40
  100. package/docs/RELEASE.md +0 -96
  101. package/docs/ROLLBACK.md +0 -9
  102. package/docs/SECURITY_REVIEW_EVIDENCE.md +0 -17
  103. package/docs/ga-readiness.json +0 -366
  104. package/docs/reports/EFFICIENCY_REVIEW.md +0 -156
  105. package/docs/reports/EXTERNAL_COMPARISON.md +0 -434
  106. package/docs/reports/FULL_HARNESS_REVIEW_2026-09-10.md +0 -200
  107. package/docs/reports/GA_BUDGET_REMEDIATION_2026-09-19.md +0 -38
  108. package/docs/reports/GA_REPAIR_RECOVERY_2026-09-19.md +0 -700
  109. package/docs/reports/GA_VALIDATION_2026-09-19.md +0 -111
  110. package/docs/reports/HARNESS_ARCHITECTURE_REVIEW_2026-09-09.md +0 -202
  111. package/docs/reports/HARNESS_AUDIT_2026-09-09.md +0 -306
  112. package/docs/reports/HARNESS_REMEDIATION_2026-09-09.md +0 -74
  113. package/docs/reports/HARNESS_REMEDIATION_2026-09-10.md +0 -90
  114. package/docs/reports/MODEL_COMPARISON_2026-09-09.md +0 -184
  115. package/docs/reports/QWEN_REMEDIATION_COMPARISON_2026-09-09.md +0 -142
  116. package/docs/reports/RC14_VALIDATION_2026-09-10.md +0 -45
  117. package/docs/reports/README.md +0 -26
  118. package/docs/reports/evidence/ga-budget-remediation-2026-09-19.json +0 -133
  119. package/docs/reports/evidence/ga-catalog-ablation-2026-09-19.json +0 -699
  120. package/docs/reports/evidence/ga-catalog-ablation-plan-2026-09-19.json +0 -40
  121. package/docs/reports/evidence/ga-catalog-observed-2026-09-19.json +0 -985
  122. package/docs/reports/evidence/ga-catalog-observed-plan-2026-09-19.json +0 -28
  123. package/docs/reports/evidence/ga-catalog-observed-status-2026-09-19.json +0 -15
  124. package/docs/reports/evidence/ga-complete-local-check-2026-09-19.json +0 -47
  125. package/docs/reports/evidence/ga-completion-reminder-2026-09-19.json +0 -22
  126. package/docs/reports/evidence/ga-fresh-cohort-plan-2026-09-19.json +0 -54
  127. package/docs/reports/evidence/ga-fresh-cohort-preparation-2026-09-19.json +0 -242
  128. package/docs/reports/evidence/ga-fresh-cohort-result-2026-09-19.json +0 -4318
  129. package/docs/reports/evidence/ga-local-validation-followup-2026-09-19.json +0 -66
  130. package/docs/reports/evidence/ga-node-consumers-2026-09-19.json +0 -31
  131. package/docs/reports/evidence/ga-package-content-validation-2026-09-19.json +0 -22
  132. package/docs/reports/evidence/ga-plan-completion-2026-09-19.json +0 -39
  133. package/docs/reports/evidence/ga-plan-reserve-2026-09-19.json +0 -573
  134. package/docs/reports/evidence/ga-plan-reserve-plan-2026-09-19.json +0 -24
  135. package/docs/reports/evidence/ga-plan-transition-2026-09-19.json +0 -19
  136. package/docs/reports/evidence/ga-private-verifier-plan-2026-09-19.json +0 -24
  137. package/docs/reports/evidence/ga-private-verifier-root-cause-2026-09-19.json +0 -41
  138. package/docs/reports/evidence/ga-published-sdk-2026-09-19.json +0 -35
  139. package/docs/reports/evidence/ga-python-checkout-imports-2026-09-19.json +0 -95
  140. package/docs/reports/evidence/ga-python-origin-delivery-2026-09-19.json +0 -211
  141. package/docs/reports/evidence/ga-python-origin-delivery-plan-2026-09-19.json +0 -27
  142. package/docs/reports/evidence/ga-qwen-patched-package-2026-09-19.json +0 -1092
  143. package/docs/reports/evidence/ga-qwen-patched-package-plan-2026-09-19.json +0 -18
  144. package/docs/reports/evidence/ga-qwen-published-2026-09-19.json +0 -571
  145. package/docs/reports/evidence/ga-qwen-published-plan-2026-09-19.json +0 -18
  146. package/docs/reports/evidence/ga-qwen-tool-choice-2026-09-19.json +0 -85
  147. package/docs/reports/evidence/ga-qwen-tool-choice-patched-2026-09-19.json +0 -85
  148. package/docs/reports/evidence/ga-recovery-development-2026-09-19.json +0 -1066
  149. package/docs/reports/evidence/ga-recovery-graded-development-2026-09-19.json +0 -1313
  150. package/docs/reports/evidence/ga-recovery-graded-v2-2026-09-19.json +0 -1151
  151. package/docs/reports/evidence/ga-recovery-v3-2026-09-19.json +0 -3943
  152. package/docs/reports/evidence/ga-recovery-v4-2026-09-19.json +0 -1055
  153. package/docs/reports/evidence/ga-recovery-v5-2026-09-19.json +0 -1245
  154. package/docs/reports/evidence/ga-recovery-v6-2026-09-19.json +0 -699
  155. package/docs/reports/evidence/ga-recovery-v7-2026-09-19.json +0 -854
  156. package/docs/reports/evidence/ga-recovery-v8-2026-09-19.json +0 -596
  157. package/docs/reports/evidence/ga-required-delivery-2026-09-19.json +0 -25
  158. package/docs/reports/evidence/ga-required-delivery-live-2026-09-19.json +0 -1337
  159. package/docs/reports/evidence/ga-required-delivery-live-plan-2026-09-19.json +0 -25
  160. package/docs/reports/evidence/ga-retained-candidate-grading-2026-09-19.json +0 -30
  161. package/docs/reports/evidence/ga-tool-selection-live-2026-09-19.json +0 -559
  162. package/docs/reports/evidence/ga-tool-selection-live-plan-2026-09-19.json +0 -28
  163. package/docs/reports/evidence/ga-tool-selection-recovery-2026-09-19.json +0 -27
  164. package/docs/reports/evidence/ga-validation-2026-09-19.json +0 -236
  165. package/docs/reports/evidence/ga-verifier-discovery-live-2026-09-19.json +0 -1369
  166. package/docs/reports/evidence/ga-verifier-discovery-plan-2026-09-19.json +0 -29
  167. package/docs/reports/evidence/ga-verifier-failure-identification-2026-09-19.json +0 -30
  168. package/docs/reports/evidence/ga-verifier-feedback-2026-09-19.json +0 -65
  169. package/docs/reports/evidence/ga-verifier-hints-live-2026-09-19.json +0 -1297
  170. package/docs/reports/evidence/ga-verifier-hints-plan-2026-09-19.json +0 -27
  171. package/docs/reports/evidence/ga-work-boundary-context-analysis-2026-09-19.json +0 -113
  172. package/docs/reports/evidence/ga-work-boundary-live-2026-09-19.json +0 -301
  173. package/docs/reports/evidence/ga-work-boundary-live-plan-2026-09-19.json +0 -84
  174. package/docs/reports/evidence/ga-work-budget-planning-2026-09-19.json +0 -32
  175. package/docs/reports/evidence/ga-working-context-2026-09-19.json +0 -22
  176. package/docs/reports/evidence/rc14-delegated-security-review-2026-09-20.json +0 -81
  177. package/docs/reports/evidence/rc14-ga-gate-2026-09-20.json +0 -18
  178. package/docs/reports/evidence/rc14-representative-2026-09-20.json +0 -509
  179. package/docs/reports/evidence/rc14-security-regressions-2026-09-20.json +0 -64
  180. package/docs/reports/evidence/security-current-2026-09-19.json +0 -91
  181. package/docs/reports/evidence/security-remediation-2026-09-19.json +0 -97
@@ -1,1570 +0,0 @@
1
- {
2
- "schemaVersion": 1,
3
- "dataset": "princeton-nlp/SWE-bench_Verified",
4
- "revision": "c104f840cc67f8b6eec6f759ebc8b2693d585d4a",
5
- "model": "qwen3.8-flash",
6
- "provider": "qwen",
7
- "reasoning": "none",
8
- "transport": "qwen-chat-completions",
9
- "limits": {
10
- "steps": 24,
11
- "outputPerTurn": 2048,
12
- "inputTokens": 100000,
13
- "outputTokens": 16000,
14
- "timeoutSeconds": 300,
15
- "memoryMb": 2048,
16
- "cpus": 2
17
- },
18
- "seed": 20260910,
19
- "repetitions": 1,
20
- "versions": {
21
- "mini-swe-agent": "2.4.6",
22
- "swebench": "4.1.0"
23
- },
24
- "implementation": {
25
- "gitCommit": "0a7b69dddaa70a54b6b5b4097691c8112d4d76eb",
26
- "sources": {
27
- "bun.lock": "d5436a51773d94333612c611a162a3e63765fec88d6bbde2a71879662b49b5cd",
28
- "evaluations/external/requirements.lock": "42e90a08f10f821ebb958af79256f707a9fb5cf365277c4f9e2715d612028179",
29
- "package.json": "ade3c3a4c0f8ee193c560637bf96f1fe2782f5bc06a42dcdbb320bd7a932a173",
30
- "scripts/benchmark-swebench.ts": "ed17063e1847017ba2e72f14732b6ff8f5b1af7cd5253b91302af75f2a9e8476",
31
- "scripts/swebench/common.py": "c19cfc30b85dc3ad8dbf1ba6a4165e72ce96f624eb1721b613cfa3207636f7be",
32
- "scripts/swebench/mini_driver.py": "bd0a77221878370ec44760122a51f07fb56d969612afa9a7f171248b2958baa6",
33
- "scripts/swebench/model-budget.ts": "67b74d45d34235586773b7cf0db4d320306b14a6f432d53c084f80563ccdfa44",
34
- "scripts/swebench/preflight.py": "aed092a75dd52f504fba6d14814b69bd627ff17d0e311a1f1279f8b613b69675",
35
- "scripts/swebench/prepare.py": "bb7edd3fba5dbbe26597c97d744c60b94dbb7bcdfd7740a4a058c8583bd27e4c",
36
- "scripts/swebench/report.py": "70afcdf0d1f081014bb75d878f3ddf9195959e96b523ddf1e0ad5d3ec8273549",
37
- "scripts/swebench/run.py": "b184258e7c32313c9c801b50d28d1a9419d0bd2b077a422012d8fc6bc449c916",
38
- "scripts/swebench/telemetry.ts": "4c6c0d58641b91539d945f341b368a9c360b6ef89ff3abb9d2f33d3cb133c12e",
39
- "scripts/swebench/test_contracts.py": "a8c1348f1dff2e563c139e6d3e83eb141b4f2a71f91a58e2c5c051ed2a431024",
40
- "scripts/swebench/zhivex-driver.ts": "7f9b9903f7e80075c4b084b5009f9fc3f8915cb9276f821ab9cb883c196e0fcb",
41
- "scripts/time-to-safe-fix-efficiency.ts": "dd20fd8839280332ae4ca80d6846e98e07ff15d79e200b206fa2ba6be2724d2f",
42
- "src/bounded-reads.ts": "9807bb460894be3279c5425af19863d1c6e419a1582ba6910ec7e7fc35703ed6",
43
- "src/capabilities.ts": "56fdb60a2f6e1d33b2dc26d38b6d2709807e491ba1268d47b97876c2c0591d65",
44
- "src/change-envelope.ts": "9047ec978c9e5117965b8093ec961d855f97bbc2c3accdc79b74cfe6a0e4f01e",
45
- "src/cli-options.ts": "97781386289e3877c1318e4c94698ca7aaf2f256b89534dfcb21e966eb7d939f",
46
- "src/cli-profiles.ts": "4fe451f1645e95f4957597a8724fa989b3c661ae1693252bfd6e74a51d20fc74",
47
- "src/cli-stream.ts": "5ab8a4b1b8cf0a3dfa7b5b1f9e4e0facf8f6e7f6c341bdf700a41dc8927c825e",
48
- "src/cli.ts": "cce6f58dfbc543acdc029b188d4394cd059d940bdd361351fd877ae2d6e94880",
49
- "src/compaction.ts": "0c4f95d2f49c5154c9b87b3f79feca18227e274a8aacd2445e387a857a37853f",
50
- "src/config.ts": "c160faaf2d9fb637c5a08a7aff65d761a12c707ed23ebb87ab0e8519e9536bf3",
51
- "src/console-context.ts": "0ad7e4559504b9ed09ab500d615dea09af1075d357f02ec3b77f003d55d951db",
52
- "src/console-input.ts": "dc8ea0cdbba1364db288d5156c5ff9819362c46b37ef35a98b0475b6c445b1f8",
53
- "src/context-engineering.ts": "1995e564f6b91974e75bf66dd421ddcbea29b1f8eee59b76eb4e34c9d6d8f47e",
54
- "src/edit-contracts.ts": "47cc3132e27f8749d54a32bbcc33bb174cf4448f0d0171b2578cdb559cf2176b",
55
- "src/errors.ts": "e02f4eb80562a83762f908bda56c950ec6a2b78725c0c8a49f41fee0057ab903",
56
- "src/execution-environment.ts": "8519e532dfafe4ae51bfbc81e0732267bc6aa9eec871ffde8c917aa001e358e5",
57
- "src/file-security.ts": "4590b46c8e4b9d3891a75cd0d0ada14e0ae7b58e8e4bdf82f68df688a1fbb750",
58
- "src/harness.ts": "00b45dfce74fd954d72e7aa5b53fe93e8983a63813f50201f4482ce3768de306",
59
- "src/index.ts": "f791311f14dc27105191d97bb5454748d10beab827da9aba3ea949ffc32fce7c",
60
- "src/json-contracts.ts": "0702e5b0ff16ff6304c74b4101bd31a3fa8eabdebe267d312b56d3871a7e302c",
61
- "src/mcp.ts": "5f3f7a35b89063264ae18ab9273b84e2e59ac8c8e611a6b64be377ba61cc1ac5",
62
- "src/operations.ts": "0775da79208c6f5d18476cc76404e84ddda610979d9d0e5dc15035aa6999613c",
63
- "src/orchestration.ts": "7383da1047363819e6560f51892fd4082e5337856a3d889cdc076f021273774e",
64
- "src/package-manager.ts": "cec4666bbab2f42e0a269709631f79f9f87c2c4dbb3cafeffacfaff761069e2c",
65
- "src/process-runtime.ts": "b7286731a0b5c96dcad0b8e3a4bf1155d0539d73303cd7ebe77ffabf858f1729",
66
- "src/providers.ts": "70ccfc7f315cd7238e83d69b719ce4908f3b620c564cbb4f00f93e102e025602",
67
- "src/replacement-edits.ts": "8a0efdd3a7a9fc3e5a5475916cbd17ef882265881bb61076db58673e80b72957",
68
- "src/routing.ts": "8e7c9f5968b4bd638cbbc86b19847ca183f3418b9a298783b96a88cffa568e22",
69
- "src/run-interruption.ts": "df99129cf2dd6b938eca629cb897c4350a7d57fe6bf0ce437f5f3f30271617df",
70
- "src/sessions.ts": "5167c8cf36bd4b8845ee278883394700321b65397f49c1dda0cf0d0de09dd06e",
71
- "src/sqlite-database.ts": "18765d1ca0c2f2ad87bfd82d311312dfe8a42006556f27bc21da5892ed3b044e",
72
- "src/state-backup.ts": "b6f8f1aedeb5dce31b65c23c7ffc50030870655b58a5aaaef9989bafb9da5c68",
73
- "src/state-directory.ts": "3711550991951e6b738a08deaa0e06da84d01864c03a9602600f592963fa3211",
74
- "src/terminal-markdown.ts": "8cffb6ebaf25108cb2bee04956d9b2973635b010f5d7ce6be21b3ceffe3e8bfd",
75
- "src/terminal-ui.ts": "e50025a8649d1f60efdf3af895b8639730ec2e86553fb80a61a776b11076c705",
76
- "src/time-to-safe-fix.ts": "4add1d49c0ea209cc496c66fa2b2012dc7bcab87e678a58c8591b3c2cdc8df8a",
77
- "src/version.ts": "b2c2f80a40adc08e8258b81e82434273f0bdfa787528b2221f14efe7f06aa747",
78
- "src/workspace.ts": "e9c2ed5c950d751d0f93a5512584d0698449b551d3270b3e115ccca380565222"
79
- }
80
- },
81
- "host": {
82
- "system": "Darwin",
83
- "architecture": "arm64",
84
- "python": "3.12.13",
85
- "docker": "29.7.2"
86
- },
87
- "environmentDifferences": [
88
- "Zhivex native OCI: read-only root and writable /workspace snapshot, operator auto-approval; mini: writable /testbed. Both deny container network and use the same derived image and CPU/memory limits.",
89
- "Zhivex tool-call limit is steps times eight; mini counts model steps and executes each response's actions. Tool counts are reported, not claimed equivalent.",
90
- "Input/output token ceilings are checked between model requests and can overshoot by one response; outputPerTurn is sent to both APIs."
91
- ],
92
- "approvalPolicy": "Automated operator approves supported isolated Zhivex tools; successful approved verification and patch import is a terminal receipt. Human wait is zero. Not a human-UX comparison.",
93
- "excludedTaskIds": [
94
- "astropy__astropy-14995",
95
- "django__django-12262",
96
- "django__django-13410",
97
- "django__django-14915",
98
- "django__django-15569",
99
- "pydata__xarray-4695",
100
- "pydata__xarray-6721",
101
- "pytest-dev__pytest-8399",
102
- "sympy__sympy-15809",
103
- "sympy__sympy-20438"
104
- ],
105
- "excludedManifestSha256": [
106
- "4902585186402c9fdd3e10db9c537109b5a108ba46b4e0806e259f6a8c4c15ab",
107
- "76739783c68673b6f659d20f757a191a52213add63e46706737535484aec6f41"
108
- ],
109
- "date": "2026-09-08",
110
- "kind": "swebench-qwen-flash-development-comparison",
111
- "tasks": [
112
- {
113
- "instance_id": "sympy__sympy-21930",
114
- "repo": "sympy/sympy",
115
- "base_commit": "de446c6d85f633271dfec1452f6f28ea783e293f",
116
- "evalImage": "swebench/sweb.eval.x86_64.sympy_1776_sympy-21930:zhx-ac9b1c65d632e5a3",
117
- "evalImageId": "sha256:ac9b1c65d632e5a3d1af240db144ff27ed9a70b5507705d50480ede6899125d1",
118
- "evalTag": "zhx-ac9b1c65d632e5a3",
119
- "image": "sha256:81c7bb0b9b74ae262830b7d4b94d042ab60c10cdbe51620c3b387262c10794fc",
120
- "checkoutHead": "b7d17e72736ba245d24b87f8f5e8c80565066de4",
121
- "baseTree": "a8e44f94a63624d1f4af136f7382125a0794eecf",
122
- "upstreamTree": "a8e44f94a63624d1f4af136f7382125a0794eecf",
123
- "imageModeDifferences": 0
124
- },
125
- {
126
- "instance_id": "scikit-learn__scikit-learn-10908",
127
- "repo": "scikit-learn/scikit-learn",
128
- "base_commit": "67d06b18c68ee4452768f8a1e868565dd4354abf",
129
- "evalImage": "swebench/sweb.eval.x86_64.scikit-learn_1776_scikit-learn-10908:zhx-8727c0f7bfb76c29",
130
- "evalImageId": "sha256:8727c0f7bfb76c294cfef16c9fe3d0dc9ddf313e7c0fb3e8e9cbc74c4955404c",
131
- "evalTag": "zhx-8727c0f7bfb76c29",
132
- "image": "sha256:4da1cd215c9209495a9aa8638026733275c766ab970311d31b1b6a8f834687d8",
133
- "checkoutHead": "de60d3cc1d489509f0d7fe1ad141171628d3d4b7",
134
- "baseTree": "d7e9da7c13c10167281372f397c7313c52c46077",
135
- "upstreamTree": "d7e9da7c13c10167281372f397c7313c52c46077",
136
- "imageModeDifferences": 0
137
- },
138
- {
139
- "instance_id": "django__django-15554",
140
- "repo": "django/django",
141
- "base_commit": "59ab3fd0e9e606d7f0f7ca26609c06ee679ece97",
142
- "evalImage": "swebench/sweb.eval.x86_64.django_1776_django-15554:zhx-9e1bc1db0e1ec55a",
143
- "evalImageId": "sha256:9e1bc1db0e1ec55a32f346c36671df7384a51a9ed84eed9b7e839e1405f04292",
144
- "evalTag": "zhx-9e1bc1db0e1ec55a",
145
- "image": "sha256:05b0c262568607baf8716c78d6bef059bf4a57a219141c6877034ee1e90a223d",
146
- "checkoutHead": "292b4e462f63a7dfa0f9cd742e14cfa0b43727d6",
147
- "baseTree": "9014f24e170c210492e65fe54bcbf236e7342c4a",
148
- "upstreamTree": "9014f24e170c210492e65fe54bcbf236e7342c4a",
149
- "imageModeDifferences": 0
150
- },
151
- {
152
- "instance_id": "scikit-learn__scikit-learn-14141",
153
- "repo": "scikit-learn/scikit-learn",
154
- "base_commit": "3d997697fdd166eff428ea9fd35734b6a8ba113e",
155
- "evalImage": "swebench/sweb.eval.x86_64.scikit-learn_1776_scikit-learn-14141:zhx-c0f6a733154dad51",
156
- "evalImageId": "sha256:c0f6a733154dad51cb8f617ca2ea09d095a6f56a1576740c2d43f9020294f2e0",
157
- "evalTag": "zhx-c0f6a733154dad51",
158
- "image": "sha256:a24afa6604e714bb6cf031b045f171e8b883e876534f941bb10ee36854612d6d",
159
- "checkoutHead": "f863dd9c48aee1eb395e5ca64e675a765cea44a9",
160
- "baseTree": "74477c0718711537b2cb5f9805ffbf9ffd072312",
161
- "upstreamTree": "74477c0718711537b2cb5f9805ffbf9ffd072312",
162
- "imageModeDifferences": 0
163
- },
164
- {
165
- "instance_id": "django__django-11265",
166
- "repo": "django/django",
167
- "base_commit": "21aa2a5e785eef1f47beb1c3760fdd7d8915ae09",
168
- "evalImage": "swebench/sweb.eval.x86_64.django_1776_django-11265:zhx-70a39003fb3ce8f0",
169
- "evalImageId": "sha256:70a39003fb3ce8f01ee82dfeb16b09bc3714ab669d3743fb3e1e5689f18ea1ef",
170
- "evalTag": "zhx-70a39003fb3ce8f0",
171
- "image": "sha256:32bc939face8eeae33506b342f5a260ab00cc49d02b616c58d7ef92b20389532",
172
- "checkoutHead": "ad2b3670a78c6db994d93f1b71bccd5d5ee4b9c4",
173
- "baseTree": "cbaf76226fd9f842b01da4d2cd52d99d2d256be2",
174
- "upstreamTree": "cbaf76226fd9f842b01da4d2cd52d99d2d256be2",
175
- "imageModeDifferences": 0
176
- }
177
- ],
178
- "report": {
179
- "schemaVersion": 1,
180
- "kind": "swebench-comparison",
181
- "manifestSha256": "fa943e995a1d4155c372bdaa66f09ffd1e64822ad13b0d4a8430e40522b16f7e",
182
- "complete": true,
183
- "candidates": {
184
- "zhivex": {
185
- "planned": 5,
186
- "recorded": 5,
187
- "resolved": 0,
188
- "resolutionRate": 0.0,
189
- "missing": 0,
190
- "gradingFailures": 0,
191
- "usageComplete": true,
192
- "totalCostUsd": null,
193
- "costPerResolvedUsd": null,
194
- "resolvedP50Ms": null
195
- },
196
- "mini-swe-agent": {
197
- "planned": 5,
198
- "recorded": 5,
199
- "resolved": 2,
200
- "resolutionRate": 0.4,
201
- "missing": 0,
202
- "gradingFailures": 0,
203
- "usageComplete": true,
204
- "totalCostUsd": null,
205
- "costPerResolvedUsd": null,
206
- "resolvedP50Ms": 40939.1438330058
207
- }
208
- },
209
- "pairedTaskResolutionDifference": -0.4,
210
- "pairedTaskBootstrap95": [
211
- -0.8,
212
- 0.0
213
- ],
214
- "evidenceBoundary": "Local matched-model comparison. Missing and failed runs remain in the denominator. Small-sample intervals are exploratory. No safety superiority is inferred from SWE-bench correctness."
215
- },
216
- "samples": [
217
- {
218
- "candidate": "zhivex",
219
- "instanceId": "sympy__sympy-21930",
220
- "repetition": 0,
221
- "resolved": false,
222
- "gradingStatus": "completed",
223
- "usageComplete": true,
224
- "costUsd": null,
225
- "status": "failed",
226
- "failure": "INPUT_TOKEN_BUDGET",
227
- "durationMs": 38459.725333,
228
- "turns": [
229
- {
230
- "index": 1,
231
- "status": "completed",
232
- "inputTokens": 3237,
233
- "outputTokens": 161,
234
- "cachedInputTokens": 0,
235
- "requestCharacters": 4987,
236
- "tools": [
237
- {
238
- "name": "list_files",
239
- "isError": true,
240
- "characters": 4,
241
- "diagnostic": {
242
- "code": "EXECUTION_FAILED",
243
- "category": "execution",
244
- "retryable": false,
245
- "chain": [
246
- {
247
- "providerReason": null,
248
- "validationIssues": [],
249
- "status": null,
250
- "errorClass": "Object",
251
- "providerHints": [],
252
- "fingerprint": "eb0264922019f3f1e60155c5264772fb2123344f17f7cded6b62a106356a4dca",
253
- "reason": null
254
- }
255
- ]
256
- }
257
- },
258
- {
259
- "name": "search_many",
260
- "isError": false,
261
- "characters": 161
262
- }
263
- ]
264
- },
265
- {
266
- "index": 2,
267
- "status": "completed",
268
- "inputTokens": 3472,
269
- "outputTokens": 55,
270
- "cachedInputTokens": 3072,
271
- "requestCharacters": 1105,
272
- "tools": [
273
- {
274
- "name": "list_files",
275
- "isError": false,
276
- "characters": 3297
277
- }
278
- ]
279
- },
280
- {
281
- "index": 3,
282
- "status": "completed",
283
- "inputTokens": 4610,
284
- "outputTokens": 84,
285
- "cachedInputTokens": 3328,
286
- "requestCharacters": 3646,
287
- "tools": [
288
- {
289
- "name": "search_many",
290
- "isError": false,
291
- "characters": 7904
292
- }
293
- ]
294
- },
295
- {
296
- "index": 4,
297
- "status": "completed",
298
- "inputTokens": 8861,
299
- "outputTokens": 255,
300
- "cachedInputTokens": 4096,
301
- "requestCharacters": 8349,
302
- "tools": [
303
- {
304
- "name": "read_files",
305
- "isError": false,
306
- "characters": 8315
307
- }
308
- ]
309
- },
310
- {
311
- "index": 5,
312
- "status": "completed",
313
- "inputTokens": 12551,
314
- "outputTokens": 114,
315
- "cachedInputTokens": 8320,
316
- "requestCharacters": 9113,
317
- "tools": [
318
- {
319
- "name": "search_many",
320
- "isError": false,
321
- "characters": 740
322
- }
323
- ]
324
- },
325
- {
326
- "index": 6,
327
- "status": "completed",
328
- "inputTokens": 13021,
329
- "outputTokens": 89,
330
- "cachedInputTokens": 12288,
331
- "requestCharacters": 1254,
332
- "tools": [
333
- {
334
- "name": "read_files",
335
- "isError": false,
336
- "characters": 6612
337
- }
338
- ]
339
- },
340
- {
341
- "index": 7,
342
- "status": "waiting_approval",
343
- "inputTokens": 15447,
344
- "outputTokens": 141,
345
- "cachedInputTokens": 12288,
346
- "requestCharacters": 7046,
347
- "tools": []
348
- },
349
- {
350
- "index": 8,
351
- "status": "waiting_approval",
352
- "inputTokens": 3762,
353
- "outputTokens": 206,
354
- "cachedInputTokens": 2048,
355
- "requestCharacters": 7519,
356
- "tools": [
357
- {
358
- "name": "run_environment_shell",
359
- "isError": false,
360
- "characters": 745,
361
- "exitCode": 128,
362
- "timedOut": false
363
- },
364
- {
365
- "name": "run_environment_shell",
366
- "isError": false,
367
- "characters": 1034,
368
- "exitCode": 0,
369
- "timedOut": false
370
- }
371
- ]
372
- },
373
- {
374
- "index": 9,
375
- "status": "completed",
376
- "inputTokens": 4403,
377
- "outputTokens": 133,
378
- "cachedInputTokens": 3328,
379
- "requestCharacters": 332,
380
- "tools": [
381
- {
382
- "name": "search_many",
383
- "isError": false,
384
- "characters": 11910
385
- }
386
- ]
387
- },
388
- {
389
- "index": 10,
390
- "status": "completed",
391
- "inputTokens": 10675,
392
- "outputTokens": 85,
393
- "cachedInputTokens": 4096,
394
- "requestCharacters": 21880,
395
- "tools": [
396
- {
397
- "name": "read_files",
398
- "isError": false,
399
- "characters": 9900
400
- }
401
- ]
402
- },
403
- {
404
- "index": 11,
405
- "status": "completed",
406
- "inputTokens": 14094,
407
- "outputTokens": 108,
408
- "cachedInputTokens": 10240,
409
- "requestCharacters": 32213,
410
- "tools": [
411
- {
412
- "name": "search_many",
413
- "isError": false,
414
- "characters": 266
415
- }
416
- ]
417
- },
418
- {
419
- "index": 12,
420
- "status": "completed",
421
- "inputTokens": 14288,
422
- "outputTokens": 98,
423
- "cachedInputTokens": 13312,
424
- "requestCharacters": 33000,
425
- "tools": [
426
- {
427
- "name": "search_many",
428
- "isError": false,
429
- "characters": 5179
430
- }
431
- ]
432
- }
433
- ],
434
- "inputTokens": 108421,
435
- "outputTokens": 1529,
436
- "cachedInputTokens": 76416,
437
- "modelCalls": 12,
438
- "compactions": [
439
- {
440
- "beforeStep": 8,
441
- "estimatedTokensBefore": 9221,
442
- "estimatedTokensAfter": 1797
443
- }
444
- ],
445
- "terminalDiagnostic": {
446
- "code": "EXECUTION_FAILED",
447
- "category": "execution",
448
- "retryable": false,
449
- "chain": [
450
- {
451
- "providerReason": null,
452
- "validationIssues": [],
453
- "status": null,
454
- "errorClass": "Object",
455
- "providerHints": [],
456
- "fingerprint": "064591d1d6249f6eadfb2800eddbdfe8576a334f23f5e1215117dc19107f21f4",
457
- "reason": "budget"
458
- }
459
- ]
460
- },
461
- "thrownDiagnostic": {
462
- "code": "EXECUTION_FAILED",
463
- "category": "execution",
464
- "retryable": false,
465
- "chain": [
466
- {
467
- "providerReason": null,
468
- "validationIssues": [],
469
- "status": null,
470
- "errorClass": "HarnessExecutionError",
471
- "providerHints": [],
472
- "fingerprint": "064591d1d6249f6eadfb2800eddbdfe8576a334f23f5e1215117dc19107f21f4",
473
- "reason": "budget"
474
- },
475
- {
476
- "providerReason": null,
477
- "validationIssues": [],
478
- "status": null,
479
- "errorClass": "Error",
480
- "providerHints": [],
481
- "fingerprint": "064591d1d6249f6eadfb2800eddbdfe8576a334f23f5e1215117dc19107f21f4",
482
- "reason": "budget"
483
- }
484
- ]
485
- },
486
- "guardrails": [],
487
- "toolCalls": 13,
488
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
489
- "protectedFilesChanged": [],
490
- "gradingMethod": "empty-patch-no-solution",
491
- "officialResolved": false,
492
- "gradingDurationMs": 0.004791014362126589,
493
- "totalDurationMs": 40163.06795802666
494
- },
495
- {
496
- "candidate": "mini-swe-agent",
497
- "instanceId": "sympy__sympy-21930",
498
- "repetition": 0,
499
- "resolved": false,
500
- "gradingStatus": "completed",
501
- "usageComplete": true,
502
- "costUsd": null,
503
- "status": "LimitsExceeded",
504
- "failure": null,
505
- "durationMs": 60712.348208006006,
506
- "toolCalls": 21,
507
- "inputTokens": 103566,
508
- "outputTokens": 2628,
509
- "cachedInputTokens": 87168,
510
- "modelCalls": 21,
511
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
512
- "protectedFilesChanged": [],
513
- "gradingMethod": "empty-patch-no-solution",
514
- "officialResolved": false,
515
- "gradingDurationMs": 0.004333036486059427,
516
- "totalDurationMs": 62332.5775419944
517
- },
518
- {
519
- "candidate": "mini-swe-agent",
520
- "instanceId": "scikit-learn__scikit-learn-10908",
521
- "repetition": 0,
522
- "resolved": true,
523
- "gradingStatus": "completed",
524
- "usageComplete": true,
525
- "costUsd": null,
526
- "status": "Submitted",
527
- "failure": null,
528
- "durationMs": 25978.861292009242,
529
- "toolCalls": 9,
530
- "inputTokens": 30440,
531
- "outputTokens": 948,
532
- "cachedInputTokens": 24704,
533
- "modelCalls": 9,
534
- "patchSha256": "5bbbc8e6f3953adf89c65636a8d6e75973081d8a28cbf3264485fabeb672de11",
535
- "protectedFilesChanged": [],
536
- "officialResolved": true,
537
- "gradingDurationMs": 20167.721583973616,
538
- "totalDurationMs": 47685.66404195735
539
- },
540
- {
541
- "candidate": "zhivex",
542
- "instanceId": "scikit-learn__scikit-learn-10908",
543
- "repetition": 0,
544
- "resolved": false,
545
- "gradingStatus": "completed",
546
- "usageComplete": true,
547
- "costUsd": null,
548
- "status": "failed",
549
- "failure": "EXECUTION_FAILED",
550
- "durationMs": 14930.337209,
551
- "turns": [
552
- {
553
- "index": 1,
554
- "status": "completed",
555
- "inputTokens": 3555,
556
- "outputTokens": 151,
557
- "cachedInputTokens": 2048,
558
- "requestCharacters": 2,
559
- "tools": [
560
- {
561
- "name": "list_files",
562
- "isError": false,
563
- "characters": 2136
564
- },
565
- {
566
- "name": "search_many",
567
- "isError": false,
568
- "characters": 305
569
- }
570
- ]
571
- },
572
- {
573
- "index": 2,
574
- "status": "completed",
575
- "inputTokens": 4507,
576
- "outputTokens": 207,
577
- "cachedInputTokens": 3328,
578
- "requestCharacters": 2,
579
- "tools": [
580
- {
581
- "name": "read_files",
582
- "isError": false,
583
- "characters": 7651
584
- },
585
- {
586
- "name": "search_many",
587
- "isError": false,
588
- "characters": 1196
589
- }
590
- ]
591
- },
592
- {
593
- "index": 3,
594
- "status": "completed",
595
- "inputTokens": 7812,
596
- "outputTokens": 86,
597
- "cachedInputTokens": 4096,
598
- "requestCharacters": 2,
599
- "tools": [
600
- {
601
- "name": "read_files",
602
- "isError": false,
603
- "characters": 6930
604
- }
605
- ]
606
- },
607
- {
608
- "index": 4,
609
- "status": "completed",
610
- "inputTokens": 10040,
611
- "outputTokens": 113,
612
- "cachedInputTokens": 7168,
613
- "requestCharacters": 2,
614
- "tools": [
615
- {
616
- "name": "search_many",
617
- "isError": false,
618
- "characters": 3123
619
- }
620
- ]
621
- }
622
- ],
623
- "inputTokens": 37456,
624
- "outputTokens": 661,
625
- "cachedInputTokens": 26624,
626
- "modelCalls": 5,
627
- "compactions": [],
628
- "terminalDiagnostic": {
629
- "code": "EXECUTION_FAILED",
630
- "category": "execution",
631
- "retryable": false,
632
- "chain": [
633
- {
634
- "providerReason": null,
635
- "validationIssues": [
636
- {
637
- "code": "invalid_type",
638
- "path": [
639
- "files",
640
- 0,
641
- "endLine"
642
- ]
643
- }
644
- ],
645
- "status": null,
646
- "errorClass": "Object",
647
- "providerHints": [],
648
- "fingerprint": "5df6c5850a252365ead3f2df4e2d069515e75359a797e6a0a64c1a1cce54424b",
649
- "reason": "invalid input"
650
- }
651
- ]
652
- },
653
- "thrownDiagnostic": {
654
- "code": "EXECUTION_FAILED",
655
- "category": "execution",
656
- "retryable": false,
657
- "chain": [
658
- {
659
- "providerReason": null,
660
- "validationIssues": [
661
- {
662
- "code": "invalid_type",
663
- "path": [
664
- "files",
665
- 0,
666
- "endLine"
667
- ]
668
- }
669
- ],
670
- "status": null,
671
- "errorClass": "HarnessExecutionError",
672
- "providerHints": [],
673
- "fingerprint": "5df6c5850a252365ead3f2df4e2d069515e75359a797e6a0a64c1a1cce54424b",
674
- "reason": "invalid input"
675
- },
676
- {
677
- "providerReason": null,
678
- "validationIssues": [
679
- {
680
- "code": "invalid_type",
681
- "path": [
682
- "files",
683
- 0,
684
- "endLine"
685
- ]
686
- }
687
- ],
688
- "status": null,
689
- "errorClass": "ValidationError",
690
- "providerHints": [],
691
- "fingerprint": "5df6c5850a252365ead3f2df4e2d069515e75359a797e6a0a64c1a1cce54424b",
692
- "reason": "invalid input"
693
- }
694
- ]
695
- },
696
- "guardrails": [],
697
- "toolCalls": 6,
698
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
699
- "protectedFilesChanged": [],
700
- "gradingMethod": "empty-patch-no-solution",
701
- "officialResolved": false,
702
- "gradingDurationMs": 0.004916044417768717,
703
- "totalDurationMs": 16330.313333019149
704
- },
705
- {
706
- "candidate": "zhivex",
707
- "instanceId": "django__django-15554",
708
- "repetition": 0,
709
- "resolved": false,
710
- "gradingStatus": "completed",
711
- "usageComplete": true,
712
- "costUsd": null,
713
- "status": "failed",
714
- "failure": "INPUT_TOKEN_BUDGET",
715
- "durationMs": 53471.117042000005,
716
- "turns": [
717
- {
718
- "index": 1,
719
- "status": "completed",
720
- "inputTokens": 3377,
721
- "outputTokens": 126,
722
- "cachedInputTokens": 2048,
723
- "requestCharacters": 2,
724
- "tools": [
725
- {
726
- "name": "list_files",
727
- "isError": false,
728
- "characters": 2164
729
- },
730
- {
731
- "name": "search_many",
732
- "isError": false,
733
- "characters": 982
734
- }
735
- ]
736
- },
737
- {
738
- "index": 2,
739
- "status": "completed",
740
- "inputTokens": 4589,
741
- "outputTokens": 56,
742
- "cachedInputTokens": 3072,
743
- "requestCharacters": 2,
744
- "tools": [
745
- {
746
- "name": "read_files",
747
- "isError": false,
748
- "characters": 7479
749
- }
750
- ]
751
- },
752
- {
753
- "index": 3,
754
- "status": "completed",
755
- "inputTokens": 6829,
756
- "outputTokens": 56,
757
- "cachedInputTokens": 4096,
758
- "requestCharacters": 2,
759
- "tools": [
760
- {
761
- "name": "read_files",
762
- "isError": false,
763
- "characters": 8009
764
- }
765
- ]
766
- },
767
- {
768
- "index": 4,
769
- "status": "completed",
770
- "inputTokens": 9105,
771
- "outputTokens": 56,
772
- "cachedInputTokens": 6656,
773
- "requestCharacters": 2,
774
- "tools": [
775
- {
776
- "name": "read_files",
777
- "isError": false,
778
- "characters": 4044
779
- }
780
- ]
781
- },
782
- {
783
- "index": 5,
784
- "status": "completed",
785
- "inputTokens": 10391,
786
- "outputTokens": 113,
787
- "cachedInputTokens": 8320,
788
- "requestCharacters": 2,
789
- "tools": [
790
- {
791
- "name": "search_many",
792
- "isError": false,
793
- "characters": 2995
794
- }
795
- ]
796
- },
797
- {
798
- "index": 6,
799
- "status": "completed",
800
- "inputTokens": 11885,
801
- "outputTokens": 54,
802
- "cachedInputTokens": 10240,
803
- "requestCharacters": 2,
804
- "tools": [
805
- {
806
- "name": "read_files",
807
- "isError": false,
808
- "characters": 7791
809
- }
810
- ]
811
- },
812
- {
813
- "index": 7,
814
- "status": "completed",
815
- "inputTokens": 5752,
816
- "outputTokens": 83,
817
- "cachedInputTokens": 2048,
818
- "requestCharacters": 13872,
819
- "tools": [
820
- {
821
- "name": "search_many",
822
- "isError": false,
823
- "characters": 8545
824
- }
825
- ]
826
- },
827
- {
828
- "index": 8,
829
- "status": "completed",
830
- "inputTokens": 9864,
831
- "outputTokens": 87,
832
- "cachedInputTokens": 5120,
833
- "requestCharacters": 332,
834
- "tools": [
835
- {
836
- "name": "read_files",
837
- "isError": false,
838
- "characters": 7764
839
- }
840
- ]
841
- },
842
- {
843
- "index": 9,
844
- "status": "completed",
845
- "inputTokens": 12349,
846
- "outputTokens": 87,
847
- "cachedInputTokens": 9216,
848
- "requestCharacters": 31423,
849
- "tools": [
850
- {
851
- "name": "read_files",
852
- "isError": false,
853
- "characters": 4380
854
- }
855
- ]
856
- },
857
- {
858
- "index": 10,
859
- "status": "completed",
860
- "inputTokens": 13874,
861
- "outputTokens": 100,
862
- "cachedInputTokens": 11264,
863
- "requestCharacters": 36250,
864
- "tools": [
865
- {
866
- "name": "search_many",
867
- "isError": false,
868
- "characters": 823
869
- }
870
- ]
871
- },
872
- {
873
- "index": 11,
874
- "status": "completed",
875
- "inputTokens": 14350,
876
- "outputTokens": 56,
877
- "cachedInputTokens": 13312,
878
- "requestCharacters": 37583,
879
- "tools": [
880
- {
881
- "name": "read_files",
882
- "isError": false,
883
- "characters": 3120
884
- }
885
- ]
886
- }
887
- ],
888
- "inputTokens": 102365,
889
- "outputTokens": 874,
890
- "cachedInputTokens": 75392,
891
- "modelCalls": 11,
892
- "compactions": [
893
- {
894
- "beforeStep": 7,
895
- "estimatedTokensBefore": 10476,
896
- "estimatedTokensAfter": 3468
897
- },
898
- {
899
- "beforeStep": 12,
900
- "estimatedTokensBefore": 10187,
901
- "estimatedTokensAfter": 2309
902
- }
903
- ],
904
- "terminalDiagnostic": {
905
- "code": "EXECUTION_FAILED",
906
- "category": "execution",
907
- "retryable": false,
908
- "chain": [
909
- {
910
- "providerReason": null,
911
- "validationIssues": [],
912
- "status": null,
913
- "errorClass": "Object",
914
- "providerHints": [],
915
- "fingerprint": "064591d1d6249f6eadfb2800eddbdfe8576a334f23f5e1215117dc19107f21f4",
916
- "reason": "budget"
917
- }
918
- ]
919
- },
920
- "thrownDiagnostic": {
921
- "code": "EXECUTION_FAILED",
922
- "category": "execution",
923
- "retryable": false,
924
- "chain": [
925
- {
926
- "providerReason": null,
927
- "validationIssues": [],
928
- "status": null,
929
- "errorClass": "HarnessExecutionError",
930
- "providerHints": [],
931
- "fingerprint": "064591d1d6249f6eadfb2800eddbdfe8576a334f23f5e1215117dc19107f21f4",
932
- "reason": "budget"
933
- },
934
- {
935
- "providerReason": null,
936
- "validationIssues": [],
937
- "status": null,
938
- "errorClass": "Error",
939
- "providerHints": [],
940
- "fingerprint": "064591d1d6249f6eadfb2800eddbdfe8576a334f23f5e1215117dc19107f21f4",
941
- "reason": "budget"
942
- }
943
- ]
944
- },
945
- "guardrails": [],
946
- "toolCalls": 12,
947
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
948
- "protectedFilesChanged": [],
949
- "gradingMethod": "empty-patch-no-solution",
950
- "officialResolved": false,
951
- "gradingDurationMs": 0.006249989382922649,
952
- "totalDurationMs": 58832.29600003688
953
- },
954
- {
955
- "candidate": "mini-swe-agent",
956
- "instanceId": "django__django-15554",
957
- "repetition": 0,
958
- "resolved": false,
959
- "gradingStatus": "completed",
960
- "usageComplete": true,
961
- "costUsd": null,
962
- "status": "LimitsExceeded",
963
- "failure": null,
964
- "durationMs": 42561.118874989916,
965
- "toolCalls": 16,
966
- "inputTokens": 105486,
967
- "outputTokens": 1510,
968
- "cachedInputTokens": 86784,
969
- "modelCalls": 16,
970
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
971
- "protectedFilesChanged": [],
972
- "gradingMethod": "empty-patch-no-solution",
973
- "officialResolved": false,
974
- "gradingDurationMs": 0.002333021257072687,
975
- "totalDurationMs": 44094.71937501803
976
- },
977
- {
978
- "candidate": "mini-swe-agent",
979
- "instanceId": "scikit-learn__scikit-learn-14141",
980
- "repetition": 0,
981
- "resolved": true,
982
- "gradingStatus": "completed",
983
- "usageComplete": true,
984
- "costUsd": null,
985
- "status": "Submitted",
986
- "failure": null,
987
- "durationMs": 19550.653749960475,
988
- "toolCalls": 6,
989
- "inputTokens": 14157,
990
- "outputTokens": 442,
991
- "cachedInputTokens": 8192,
992
- "modelCalls": 6,
993
- "patchSha256": "b1ca3d45bf4e8d854c1921ac8f47ff519d0987ef33adc3c787e2a2c68bb2aba4",
994
- "protectedFilesChanged": [],
995
- "officialResolved": true,
996
- "gradingDurationMs": 19867.57433298044,
997
- "totalDurationMs": 40939.1438330058
998
- },
999
- {
1000
- "candidate": "zhivex",
1001
- "instanceId": "scikit-learn__scikit-learn-14141",
1002
- "repetition": 0,
1003
- "resolved": false,
1004
- "gradingStatus": "completed",
1005
- "usageComplete": true,
1006
- "costUsd": null,
1007
- "status": "failed",
1008
- "failure": "EXECUTION_FAILED",
1009
- "durationMs": 38582.481708,
1010
- "turns": [
1011
- {
1012
- "index": 1,
1013
- "status": "completed",
1014
- "inputTokens": 3076,
1015
- "outputTokens": 132,
1016
- "cachedInputTokens": 2048,
1017
- "requestCharacters": 4396,
1018
- "tools": [
1019
- {
1020
- "name": "list_files",
1021
- "isError": false,
1022
- "characters": 2631
1023
- },
1024
- {
1025
- "name": "search_many",
1026
- "isError": false,
1027
- "characters": 2714
1028
- }
1029
- ]
1030
- },
1031
- {
1032
- "index": 2,
1033
- "status": "completed",
1034
- "inputTokens": 5320,
1035
- "outputTokens": 60,
1036
- "cachedInputTokens": 3072,
1037
- "requestCharacters": 6143,
1038
- "tools": [
1039
- {
1040
- "name": "read_files",
1041
- "isError": false,
1042
- "characters": 7000
1043
- }
1044
- ]
1045
- },
1046
- {
1047
- "index": 3,
1048
- "status": "waiting_approval",
1049
- "inputTokens": 7907,
1050
- "outputTokens": 173,
1051
- "cachedInputTokens": 5120,
1052
- "requestCharacters": 7427,
1053
- "tools": []
1054
- },
1055
- {
1056
- "index": 4,
1057
- "status": "waiting_approval",
1058
- "inputTokens": 8858,
1059
- "outputTokens": 197,
1060
- "cachedInputTokens": 7168,
1061
- "requestCharacters": 20648,
1062
- "tools": [
1063
- {
1064
- "name": "run_environment_command",
1065
- "isError": false,
1066
- "characters": 394,
1067
- "exitCode": 0,
1068
- "timedOut": false
1069
- }
1070
- ]
1071
- },
1072
- {
1073
- "index": 5,
1074
- "status": "waiting_approval",
1075
- "inputTokens": 9362,
1076
- "outputTokens": 179,
1077
- "cachedInputTokens": 8320,
1078
- "requestCharacters": 21849,
1079
- "tools": [
1080
- {
1081
- "name": "apply_reviewed_replacement",
1082
- "isError": false,
1083
- "characters": 475
1084
- }
1085
- ]
1086
- },
1087
- {
1088
- "index": 6,
1089
- "status": "waiting_approval",
1090
- "inputTokens": 9857,
1091
- "outputTokens": 117,
1092
- "cachedInputTokens": 9216,
1093
- "requestCharacters": 22999,
1094
- "tools": [
1095
- {
1096
- "name": "apply_reviewed_replacement",
1097
- "isError": false,
1098
- "characters": 485
1099
- }
1100
- ]
1101
- },
1102
- {
1103
- "index": 7,
1104
- "status": "waiting_approval",
1105
- "inputTokens": 10513,
1106
- "outputTokens": 234,
1107
- "cachedInputTokens": 9216,
1108
- "requestCharacters": 25583,
1109
- "tools": [
1110
- {
1111
- "name": "run_environment_command",
1112
- "isError": false,
1113
- "characters": 1973,
1114
- "exitCode": 1,
1115
- "timedOut": false
1116
- }
1117
- ]
1118
- },
1119
- {
1120
- "index": 8,
1121
- "status": "waiting_approval",
1122
- "inputTokens": 3951,
1123
- "outputTokens": 177,
1124
- "cachedInputTokens": 2048,
1125
- "requestCharacters": 8556,
1126
- "tools": [
1127
- {
1128
- "name": "run_environment_command",
1129
- "isError": false,
1130
- "characters": 927,
1131
- "exitCode": 0,
1132
- "timedOut": false
1133
- }
1134
- ]
1135
- },
1136
- {
1137
- "index": 9,
1138
- "status": "completed",
1139
- "inputTokens": 4436,
1140
- "outputTokens": 15,
1141
- "cachedInputTokens": 3328,
1142
- "requestCharacters": 10261,
1143
- "tools": [
1144
- {
1145
- "name": "run_environment_command",
1146
- "isError": false,
1147
- "characters": 887,
1148
- "exitCode": 0,
1149
- "timedOut": false
1150
- }
1151
- ]
1152
- },
1153
- {
1154
- "index": 10,
1155
- "status": "waiting_approval",
1156
- "inputTokens": 4932,
1157
- "outputTokens": 133,
1158
- "cachedInputTokens": 4096,
1159
- "requestCharacters": 11378,
1160
- "tools": [
1161
- {
1162
- "name": "inspect_environment_patch",
1163
- "isError": false,
1164
- "characters": 795
1165
- }
1166
- ]
1167
- }
1168
- ],
1169
- "inputTokens": 68212,
1170
- "outputTokens": 1417,
1171
- "cachedInputTokens": 53632,
1172
- "modelCalls": 10,
1173
- "compactions": [
1174
- {
1175
- "beforeStep": 8,
1176
- "estimatedTokensBefore": 6988,
1177
- "estimatedTokensAfter": 2057
1178
- }
1179
- ],
1180
- "terminalDiagnostic": null,
1181
- "thrownDiagnostic": {
1182
- "code": "EXECUTION_FAILED",
1183
- "category": "execution",
1184
- "retryable": true,
1185
- "chain": [
1186
- {
1187
- "providerReason": null,
1188
- "validationIssues": [],
1189
- "status": null,
1190
- "errorClass": "HarnessExecutionError",
1191
- "providerHints": [],
1192
- "fingerprint": "2d43809b162b8543195aa4f8f580e478018783c4e0baff12061c2cbd1496942f",
1193
- "reason": null
1194
- },
1195
- {
1196
- "providerReason": null,
1197
- "validationIssues": [],
1198
- "status": null,
1199
- "errorClass": "Error",
1200
- "providerHints": [],
1201
- "fingerprint": "2d43809b162b8543195aa4f8f580e478018783c4e0baff12061c2cbd1496942f",
1202
- "reason": null
1203
- }
1204
- ]
1205
- },
1206
- "guardrails": [],
1207
- "toolCalls": 13,
1208
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
1209
- "protectedFilesChanged": [],
1210
- "gradingMethod": "empty-patch-no-solution",
1211
- "officialResolved": false,
1212
- "gradingDurationMs": 0.005082983989268541,
1213
- "totalDurationMs": 40729.9125000136
1214
- },
1215
- {
1216
- "candidate": "zhivex",
1217
- "instanceId": "django__django-11265",
1218
- "repetition": 0,
1219
- "resolved": false,
1220
- "gradingStatus": "completed",
1221
- "usageComplete": true,
1222
- "costUsd": null,
1223
- "status": "failed",
1224
- "failure": "INPUT_TOKEN_BUDGET",
1225
- "durationMs": 55047.484833,
1226
- "turns": [
1227
- {
1228
- "index": 1,
1229
- "status": "completed",
1230
- "inputTokens": 4137,
1231
- "outputTokens": 153,
1232
- "cachedInputTokens": 2048,
1233
- "requestCharacters": 8177,
1234
- "tools": [
1235
- {
1236
- "name": "list_files",
1237
- "isError": false,
1238
- "characters": 2793
1239
- },
1240
- {
1241
- "name": "search_many",
1242
- "isError": false,
1243
- "characters": 400
1244
- }
1245
- ]
1246
- },
1247
- {
1248
- "index": 2,
1249
- "status": "completed",
1250
- "inputTokens": 5261,
1251
- "outputTokens": 154,
1252
- "cachedInputTokens": 4096,
1253
- "requestCharacters": 4075,
1254
- "tools": [
1255
- {
1256
- "name": "search_many",
1257
- "isError": false,
1258
- "characters": 13628
1259
- },
1260
- {
1261
- "name": "read_files",
1262
- "isError": false,
1263
- "characters": 6613
1264
- }
1265
- ]
1266
- },
1267
- {
1268
- "index": 3,
1269
- "status": "completed",
1270
- "inputTokens": 13547,
1271
- "outputTokens": 183,
1272
- "cachedInputTokens": 5120,
1273
- "requestCharacters": 21101,
1274
- "tools": [
1275
- {
1276
- "name": "search_many",
1277
- "isError": false,
1278
- "characters": 690
1279
- },
1280
- {
1281
- "name": "read_files",
1282
- "isError": false,
1283
- "characters": 8996
1284
- }
1285
- ]
1286
- },
1287
- {
1288
- "index": 4,
1289
- "status": "completed",
1290
- "inputTokens": 6392,
1291
- "outputTokens": 87,
1292
- "cachedInputTokens": 2048,
1293
- "requestCharacters": 16095,
1294
- "tools": [
1295
- {
1296
- "name": "read_files",
1297
- "isError": false,
1298
- "characters": 9639
1299
- }
1300
- ]
1301
- },
1302
- {
1303
- "index": 5,
1304
- "status": "completed",
1305
- "inputTokens": 9533,
1306
- "outputTokens": 83,
1307
- "cachedInputTokens": 6144,
1308
- "requestCharacters": 10087,
1309
- "tools": [
1310
- {
1311
- "name": "search_many",
1312
- "isError": false,
1313
- "characters": 2261
1314
- }
1315
- ]
1316
- },
1317
- {
1318
- "index": 6,
1319
- "status": "completed",
1320
- "inputTokens": 10623,
1321
- "outputTokens": 116,
1322
- "cachedInputTokens": 9216,
1323
- "requestCharacters": 2729,
1324
- "tools": [
1325
- {
1326
- "name": "read_files",
1327
- "isError": false,
1328
- "characters": 12569
1329
- }
1330
- ]
1331
- },
1332
- {
1333
- "index": 7,
1334
- "status": "completed",
1335
- "inputTokens": 6987,
1336
- "outputTokens": 87,
1337
- "cachedInputTokens": 3072,
1338
- "requestCharacters": 18880,
1339
- "tools": [
1340
- {
1341
- "name": "read_files",
1342
- "isError": false,
1343
- "characters": 10718
1344
- }
1345
- ]
1346
- },
1347
- {
1348
- "index": 8,
1349
- "status": "waiting_approval",
1350
- "inputTokens": 10330,
1351
- "outputTokens": 51,
1352
- "cachedInputTokens": 6144,
1353
- "requestCharacters": 11497,
1354
- "tools": []
1355
- },
1356
- {
1357
- "index": 9,
1358
- "status": "waiting_approval",
1359
- "inputTokens": 11580,
1360
- "outputTokens": 72,
1361
- "cachedInputTokens": 10240,
1362
- "requestCharacters": 35660,
1363
- "tools": [
1364
- {
1365
- "name": "run_environment_shell",
1366
- "isError": false,
1367
- "characters": 4893,
1368
- "exitCode": 0,
1369
- "timedOut": false
1370
- },
1371
- {
1372
- "name": "run_environment_shell",
1373
- "isError": false,
1374
- "characters": 5551,
1375
- "exitCode": 0,
1376
- "timedOut": false
1377
- }
1378
- ]
1379
- },
1380
- {
1381
- "index": 10,
1382
- "status": "completed",
1383
- "inputTokens": 4712,
1384
- "outputTokens": 95,
1385
- "cachedInputTokens": 3072,
1386
- "requestCharacters": 332,
1387
- "tools": [
1388
- {
1389
- "name": "read_files",
1390
- "isError": false,
1391
- "characters": 8694
1392
- }
1393
- ]
1394
- },
1395
- {
1396
- "index": 11,
1397
- "status": "completed",
1398
- "inputTokens": 7337,
1399
- "outputTokens": 98,
1400
- "cachedInputTokens": 4096,
1401
- "requestCharacters": 21403,
1402
- "tools": [
1403
- {
1404
- "name": "search_many",
1405
- "isError": false,
1406
- "characters": 1408
1407
- }
1408
- ]
1409
- },
1410
- {
1411
- "index": 12,
1412
- "status": "completed",
1413
- "inputTokens": 8031,
1414
- "outputTokens": 87,
1415
- "cachedInputTokens": 7168,
1416
- "requestCharacters": 23331,
1417
- "tools": [
1418
- {
1419
- "name": "read_files",
1420
- "isError": false,
1421
- "characters": 21195
1422
- }
1423
- ]
1424
- },
1425
- {
1426
- "index": 13,
1427
- "status": "completed",
1428
- "inputTokens": 9796,
1429
- "outputTokens": 121,
1430
- "cachedInputTokens": 3072,
1431
- "requestCharacters": 331,
1432
- "tools": [
1433
- {
1434
- "name": "search_many",
1435
- "isError": false,
1436
- "characters": 883
1437
- }
1438
- ]
1439
- }
1440
- ],
1441
- "inputTokens": 108266,
1442
- "outputTokens": 1387,
1443
- "cachedInputTokens": 65536,
1444
- "modelCalls": 13,
1445
- "compactions": [
1446
- {
1447
- "beforeStep": 4,
1448
- "estimatedTokensBefore": 10996,
1449
- "estimatedTokensAfter": 4024
1450
- },
1451
- {
1452
- "beforeStep": 7,
1453
- "estimatedTokensBefore": 10499,
1454
- "estimatedTokensAfter": 4720
1455
- },
1456
- {
1457
- "beforeStep": 10,
1458
- "estimatedTokensBefore": 10335,
1459
- "estimatedTokensAfter": 2968
1460
- },
1461
- {
1462
- "beforeStep": 13,
1463
- "estimatedTokensBefore": 11161,
1464
- "estimatedTokensAfter": 6910
1465
- }
1466
- ],
1467
- "terminalDiagnostic": {
1468
- "code": "EXECUTION_FAILED",
1469
- "category": "execution",
1470
- "retryable": false,
1471
- "chain": [
1472
- {
1473
- "providerReason": null,
1474
- "validationIssues": [],
1475
- "status": null,
1476
- "errorClass": "Object",
1477
- "providerHints": [],
1478
- "fingerprint": "064591d1d6249f6eadfb2800eddbdfe8576a334f23f5e1215117dc19107f21f4",
1479
- "reason": "budget"
1480
- }
1481
- ]
1482
- },
1483
- "thrownDiagnostic": {
1484
- "code": "EXECUTION_FAILED",
1485
- "category": "execution",
1486
- "retryable": false,
1487
- "chain": [
1488
- {
1489
- "providerReason": null,
1490
- "validationIssues": [],
1491
- "status": null,
1492
- "errorClass": "HarnessExecutionError",
1493
- "providerHints": [],
1494
- "fingerprint": "064591d1d6249f6eadfb2800eddbdfe8576a334f23f5e1215117dc19107f21f4",
1495
- "reason": "budget"
1496
- },
1497
- {
1498
- "providerReason": null,
1499
- "validationIssues": [],
1500
- "status": null,
1501
- "errorClass": "Error",
1502
- "providerHints": [],
1503
- "fingerprint": "064591d1d6249f6eadfb2800eddbdfe8576a334f23f5e1215117dc19107f21f4",
1504
- "reason": "budget"
1505
- }
1506
- ]
1507
- },
1508
- "guardrails": [],
1509
- "toolCalls": 16,
1510
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
1511
- "protectedFilesChanged": [],
1512
- "gradingMethod": "empty-patch-no-solution",
1513
- "officialResolved": false,
1514
- "gradingDurationMs": 0.00429095234721899,
1515
- "totalDurationMs": 60055.93599996064
1516
- },
1517
- {
1518
- "candidate": "mini-swe-agent",
1519
- "instanceId": "django__django-11265",
1520
- "repetition": 0,
1521
- "resolved": false,
1522
- "gradingStatus": "completed",
1523
- "usageComplete": true,
1524
- "costUsd": null,
1525
- "status": "LimitsExceeded",
1526
- "failure": null,
1527
- "durationMs": 33320.89087495115,
1528
- "toolCalls": 14,
1529
- "inputTokens": 105832,
1530
- "outputTokens": 876,
1531
- "cachedInputTokens": 87808,
1532
- "modelCalls": 14,
1533
- "patchSha256": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
1534
- "protectedFilesChanged": [],
1535
- "gradingMethod": "empty-patch-no-solution",
1536
- "officialResolved": false,
1537
- "gradingDurationMs": 0.002125045284628868,
1538
- "totalDurationMs": 34937.70324997604
1539
- }
1540
- ],
1541
- "rawSamplesSha256": "89a0afc821db1b7cf69d21f16be6c323e6730fab0af68f1e2c60b15f4c36b34d",
1542
- "validation": {
1543
- "testsPassed": 446,
1544
- "preflightPassed": 5,
1545
- "sourceHashesVerifiedAfterRun": true,
1546
- "verifiedSourceCount": 52,
1547
- "qwenStreamSmoke": {
1548
- "inputTokens": 288,
1549
- "outputTokens": 13,
1550
- "toolEvents": 1
1551
- }
1552
- },
1553
- "sdkDependencies": {
1554
- "@zhivex-ai/agents": "1.3.1",
1555
- "@zhivex-ai/core": "1.13.0",
1556
- "@zhivex-ai/gemini": "0.11.0",
1557
- "@zhivex-ai/meta": "0.2.2",
1558
- "@zhivex-ai/openai": "0.11.2",
1559
- "@zhivex-ai/qwen": "0.11.4"
1560
- },
1561
- "limitations": [
1562
- "Five known development cases, one repetition; not a new holdout.",
1563
- "Model, transport, reasoning setting and harness behavior differ from the earlier GPT run; no causal before/after efficiency or model ranking claim.",
1564
- "Native requires successful approved verification before import; mini uses its original submission flow.",
1565
- "One response may exceed a cumulative ceiling. No monetary rates supplied.",
1566
- "Earlier integration attempt with Qwen 0.11.1 was aborted because stream usage was missing; its interrupted billed usage is unknown and it is not a completed comparison.",
1567
- "Empty native patches count as unresolved without hidden test execution. Local ARM-to-AMD64 emulation is not general latency evidence."
1568
- ],
1569
- "abortedIntegrationAttempt": "results/swebench/qwen-flash-comparison-2026-09-08"
1570
- }