@wendellresearch-ai/benchwarmers 0.2.0 → 0.2.1

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 (96) hide show
  1. package/AGENTS.md +147 -147
  2. package/BENCHMARKS.md +262 -262
  3. package/GATEWAY.md +272 -272
  4. package/NOTICE +28 -28
  5. package/PRIVACY.md +46 -46
  6. package/README.md +368 -314
  7. package/SECURITY.md +83 -83
  8. package/TERMS.md +22 -22
  9. package/USAGE.md +170 -163
  10. package/VERSION +1 -1
  11. package/bin/benchwarmers.js +16 -74
  12. package/cli.py +666 -0
  13. package/examples/config/gateway.json +5 -5
  14. package/examples/config/model-roles.json +11 -11
  15. package/gateway.py +755 -755
  16. package/launcher.py +120 -0
  17. package/mcp_server.py +266 -266
  18. package/metrics/alerts.py +116 -116
  19. package/metrics/aliases.py +41 -41
  20. package/metrics/bench_collections.py +141 -141
  21. package/metrics/catalog.py +295 -295
  22. package/metrics/client.py +394 -394
  23. package/metrics/cliproxy.py +191 -191
  24. package/metrics/connect.py +691 -691
  25. package/metrics/cost.py +152 -152
  26. package/metrics/decide.py +227 -227
  27. package/metrics/discover.py +318 -318
  28. package/metrics/endpoints.py +436 -436
  29. package/metrics/guard.py +617 -592
  30. package/metrics/learn.py +221 -221
  31. package/metrics/litellm_hub.py +206 -206
  32. package/metrics/model_roles.py +84 -84
  33. package/metrics/offload.py +166 -166
  34. package/metrics/owui.py +205 -205
  35. package/metrics/paths.py +60 -60
  36. package/metrics/route.py +579 -579
  37. package/metrics/scheduler.py +275 -275
  38. package/metrics/sources.py +167 -167
  39. package/metrics/wrong.py +144 -144
  40. package/package.json +68 -65
  41. package/perf/README.md +89 -89
  42. package/perf/run.py +261 -261
  43. package/scripts/auto_retire.py +367 -367
  44. package/scripts/check.py +108 -106
  45. package/scripts/check_deprecations.py +378 -378
  46. package/scripts/check_themes.py +103 -103
  47. package/scripts/init.py +124 -124
  48. package/scripts/list_served.py +78 -78
  49. package/scripts/notice.py +78 -78
  50. package/scripts/onboard.py +401 -331
  51. package/scripts/provider_tags.py +129 -129
  52. package/scripts/reprefix.py +136 -136
  53. package/scripts/sync_config.py +105 -105
  54. package/scripts/test_auth.py +255 -255
  55. package/scripts/test_cli.py +189 -0
  56. package/scripts/test_client.py +233 -233
  57. package/scripts/test_connect.py +222 -222
  58. package/scripts/test_first_run.py +180 -180
  59. package/scripts/test_hubs.py +328 -328
  60. package/scripts/test_launcher.py +67 -0
  61. package/scripts/test_llm_log.py +85 -81
  62. package/scripts/test_mcp.py +178 -178
  63. package/scripts/test_merged.py +214 -214
  64. package/scripts/test_no_owui.py +291 -291
  65. package/scripts/test_owui.py +90 -90
  66. package/scripts/test_serve_edit.py +448 -438
  67. package/scripts/test_verdict.py +98 -98
  68. package/scripts/test_vision.py +228 -228
  69. package/serve.py +3904 -3764
  70. package/suite/mine.py +154 -154
  71. package/suite/pace.py +145 -145
  72. package/suite/run.py +1046 -1042
  73. package/suite/seed.py +421 -421
  74. package/suite/tasks/claim-alias-unbound/fixtures/onboard.py +286 -286
  75. package/suite/tasks/claim-error-text-string/fixtures/decide.py +187 -187
  76. package/suite/tasks/claim-list-body/fixtures/gateway.py +235 -235
  77. package/suite/tasks/claim-member-nominated-twice/fixtures/gateway.py +284 -284
  78. package/suite/tasks/claim-nominee-uses-slot/fixtures/gateway.py +284 -284
  79. package/suite/tasks/claim-silent-exit/fixtures/gateway.py +235 -235
  80. package/suite/tasks/claim-sse-hang/fixtures/gateway.py +235 -235
  81. package/suite/tasks/claim-unknown-nominee-crash/fixtures/gateway.py +284 -284
  82. package/suite/tasks/rebench-matchms-809-bug/task.json +23 -23
  83. package/suite/tasks/rebench-matchms-809-control/task.json +27 -27
  84. package/suite/tasks/rebench-more-itertools-1028-bug/task.json +23 -23
  85. package/suite/tasks/rebench-more-itertools-1028-control/task.json +27 -27
  86. package/suite/tasks/rebench-narwhals-2864-bug/task.json +23 -23
  87. package/suite/tasks/rebench-narwhals-2864-control/task.json +27 -27
  88. package/suite/tasks/rebench-quantecon-py-769-bug/task.json +23 -23
  89. package/suite/tasks/rebench-quantecon-py-769-control/task.json +27 -27
  90. package/suite/tasks/rebench-scrapy-6867-bug/task.json +23 -23
  91. package/suite/tasks/rebench-scrapy-6867-control/task.json +27 -27
  92. package/ui/bench.css +430 -425
  93. package/ui/bench.js +1756 -1576
  94. package/ui/themes/dark.css +35 -35
  95. package/ui/themes/hardwood.css +52 -52
  96. package/ui/themes/rainforest.css +32 -32
package/scripts/notice.py CHANGED
@@ -1,78 +1,78 @@
1
- """Write NOTICE from what the repository actually ships.
2
-
3
- python scripts/notice.py # rewrite NOTICE
4
- python scripts/notice.py --check # fail if NOTICE is out of date (scripts/check.py runs this)
5
-
6
- RELEASE-PLAN.md Phase 4: the credits are generated, not kept by hand. Each
7
- public real-bug task records the project it came from, its licence and the
8
- pull request (`upstream`, `license`, `source_url` in task.json), so adding a
9
- task adds its credit and dropping one removes it. The fonts and the dataset
10
- are listed from what is in the tree.
11
-
12
- Still to do before release (Phase 7): the full licence texts the projects'
13
- licences require (Apache-2.0, BSD-3-Clause with its copyright lines) in a
14
- LICENSES/ folder.
15
- """
16
-
17
- from __future__ import annotations
18
-
19
- import collections
20
- import json
21
- import pathlib
22
- import sys
23
-
24
- ROOT = pathlib.Path(__file__).resolve().parent.parent
25
- NOTICE = ROOT / "NOTICE"
26
- DATASET = "nebius/SWE-rebench-V2" # suite/rebench.py DATASET
27
-
28
-
29
- def render() -> str:
30
- projects: dict[tuple[str, str], list[str]] = collections.defaultdict(list)
31
- for path in sorted((ROOT / "suite" / "tasks").glob("rebench-*/task.json")):
32
- spec = json.loads(path.read_text(encoding="utf-8"))
33
- key = (spec.get("upstream") or "unknown", spec.get("license") or "unknown licence")
34
- if spec.get("source_url") and spec["source_url"] not in projects[key]:
35
- projects[key].append(spec["source_url"])
36
- lines = [
37
- "Benchwarmers, by Wendell Research",
38
- "Copyright (c) 2026 Wendell Research. Licensed under the MIT License (LICENSE).",
39
- "",
40
- "This repository also contains material from others, under their own licences:",
41
- "",
42
- ]
43
- if projects:
44
- lines += [f"Real-bug tasks (suite/tasks/rebench-*) are drawn from the {DATASET} dataset",
45
- "on Hugging Face (CC-BY-4.0). Each task holds code from the project it names,",
46
- "under that project's licence:", ""]
47
- for (upstream, licence), urls in sorted(projects.items(), key=lambda kv: kv[0][0].lower()):
48
- lines.append(f" {upstream} ({licence})")
49
- lines += [f" {u}" for u in urls]
50
- lines.append("")
51
- if sorted((ROOT / "ui" / "fonts").glob("chivo*.woff2")):
52
- lines += ["Fonts (ui/fonts): Chivo and Chivo Mono, Copyright The Chivo Project Authors",
53
- "(https://github.com/Omnibus-Type/Chivo), under the SIL Open Font License 1.1",
54
- "(ui/fonts/OFL-*.txt).", ""]
55
- if (ROOT / "ui" / "mascot.png").exists():
56
- lines += ["The Wendell Research mascot (ui/mascot.png, drawn at centre court in the",
57
- "Hardwood theme) is a registered trademark of Wendell Research. It is not",
58
- "covered by the MIT License: a fork may keep it in place unchanged, but should",
59
- "not use it to name or brand another product.", ""]
60
- return "\n".join(lines)
61
-
62
-
63
- def main() -> int:
64
- text = render()
65
- if "--check" in sys.argv:
66
- current = NOTICE.read_text(encoding="utf-8") if NOTICE.exists() else ""
67
- if current.replace("\r\n", "\n") != text:
68
- print("NOTICE is out of date: run python scripts/notice.py")
69
- return 1
70
- print("NOTICE is current")
71
- return 0
72
- NOTICE.write_text(text, encoding="utf-8", newline="\n")
73
- print(f"wrote {NOTICE.name}: {text.count(chr(10))} lines")
74
- return 0
75
-
76
-
77
- if __name__ == "__main__":
78
- sys.exit(main())
1
+ """Write NOTICE from what the repository actually ships.
2
+
3
+ python scripts/notice.py # rewrite NOTICE
4
+ python scripts/notice.py --check # fail if NOTICE is out of date (scripts/check.py runs this)
5
+
6
+ RELEASE-PLAN.md Phase 4: the credits are generated, not kept by hand. Each
7
+ public real-bug task records the project it came from, its licence and the
8
+ pull request (`upstream`, `license`, `source_url` in task.json), so adding a
9
+ task adds its credit and dropping one removes it. The fonts and the dataset
10
+ are listed from what is in the tree.
11
+
12
+ Still to do before release (Phase 7): the full licence texts the projects'
13
+ licences require (Apache-2.0, BSD-3-Clause with its copyright lines) in a
14
+ LICENSES/ folder.
15
+ """
16
+
17
+ from __future__ import annotations
18
+
19
+ import collections
20
+ import json
21
+ import pathlib
22
+ import sys
23
+
24
+ ROOT = pathlib.Path(__file__).resolve().parent.parent
25
+ NOTICE = ROOT / "NOTICE"
26
+ DATASET = "nebius/SWE-rebench-V2" # suite/rebench.py DATASET
27
+
28
+
29
+ def render() -> str:
30
+ projects: dict[tuple[str, str], list[str]] = collections.defaultdict(list)
31
+ for path in sorted((ROOT / "suite" / "tasks").glob("rebench-*/task.json")):
32
+ spec = json.loads(path.read_text(encoding="utf-8"))
33
+ key = (spec.get("upstream") or "unknown", spec.get("license") or "unknown licence")
34
+ if spec.get("source_url") and spec["source_url"] not in projects[key]:
35
+ projects[key].append(spec["source_url"])
36
+ lines = [
37
+ "Benchwarmers, by Wendell Research",
38
+ "Copyright (c) 2026 Wendell Research. Licensed under the MIT License (LICENSE).",
39
+ "",
40
+ "This repository also contains material from others, under their own licences:",
41
+ "",
42
+ ]
43
+ if projects:
44
+ lines += [f"Real-bug tasks (suite/tasks/rebench-*) are drawn from the {DATASET} dataset",
45
+ "on Hugging Face (CC-BY-4.0). Each task holds code from the project it names,",
46
+ "under that project's licence:", ""]
47
+ for (upstream, licence), urls in sorted(projects.items(), key=lambda kv: kv[0][0].lower()):
48
+ lines.append(f" {upstream} ({licence})")
49
+ lines += [f" {u}" for u in urls]
50
+ lines.append("")
51
+ if sorted((ROOT / "ui" / "fonts").glob("chivo*.woff2")):
52
+ lines += ["Fonts (ui/fonts): Chivo and Chivo Mono, Copyright The Chivo Project Authors",
53
+ "(https://github.com/Omnibus-Type/Chivo), under the SIL Open Font License 1.1",
54
+ "(ui/fonts/OFL-*.txt).", ""]
55
+ if (ROOT / "ui" / "mascot.png").exists():
56
+ lines += ["The Wendell Research mascot (ui/mascot.png, drawn at centre court in the",
57
+ "Hardwood theme) is a registered trademark of Wendell Research. It is not",
58
+ "covered by the MIT License: a fork may keep it in place unchanged, but should",
59
+ "not use it to name or brand another product.", ""]
60
+ return "\n".join(lines)
61
+
62
+
63
+ def main() -> int:
64
+ text = render()
65
+ if "--check" in sys.argv:
66
+ current = NOTICE.read_text(encoding="utf-8") if NOTICE.exists() else ""
67
+ if current.replace("\r\n", "\n") != text:
68
+ print("NOTICE is out of date: run python scripts/notice.py")
69
+ return 1
70
+ print("NOTICE is current")
71
+ return 0
72
+ NOTICE.write_text(text, encoding="utf-8", newline="\n")
73
+ print(f"wrote {NOTICE.name}: {text.count(chr(10))} lines")
74
+ return 0
75
+
76
+
77
+ if __name__ == "__main__":
78
+ sys.exit(main())