Google Git Google Git
Sign in
apache/tooling-llmao/HEAD
  1. 518e597 Merge pull request #47 from apache/dependabot/uv/multidict-6.9.1 by Greg Stein · 10 hours ago main
  2. b31fd9c Merge pull request #53 from apache/dependabot/uv/aiohttp-3.14.4 by Greg Stein · 10 hours ago
  3. 79404b3 Merge pull request #48 from apache/dependabot/uv/werkzeug-3.1.9 by Greg Stein · 10 hours ago
  4. 4ee0d51 tests: license header on test_trust_store.py by Greg Stein · 12 hours ago
  5. 371a535 build(deps): bump multidict from 6.7.1 to 6.9.1 by dependabot[bot] · 20 hours ago
  6. 0d67957 build(deps): bump aiohttp from 3.14.3 to 3.14.4 by dependabot[bot] · 20 hours ago
  7. 11eb778 build(deps): bump werkzeug from 3.1.8 to 3.1.9 by dependabot[bot] · 20 hours ago
  8. 92deba6 docs: status note for 22 October by Greg Stein · 29 hours ago
  9. 6887499 deps: certifi 2026.10.19, and a test that keeps a distrusted root out by Andrew Musselman · 2 days ago
  10. 328749b docs: box-protocol.md replaces vllm-fleet-design.md, as fact not proposal by Greg Stein · 2 days ago
  11. 777b9ea docs: remove fleet-state.md by Greg Stein · 2 days ago
  12. 22d8acd docs: the reconcile design is a working document, meant to become obsolete by Greg Stein · 2 days ago
  13. 06a921c docs: fleet reconcile, the stamp backfilled and the hand check's findings by Greg Stein · 2 days ago
  14. 08815b4 reconcile: the key's display form, used by every log line and page cell by Greg Stein · 2 days ago
  15. 52818a2 deployments: stamp the key on every /model/new, and look ids up by it by Greg Stein · 2 days ago
  16. 7bd46bc deployments: re-read the LiteLLM deployments after every write by Greg Stein · 2 days ago
  17. 039c403 fleet: one working-key rule, shared by the deployment body and /v1/models by Greg Stein · 2 days ago
  18. b9b5ce3 fleet: ORPHAN_PASSES for the skew clock, UNLISTED_PASSES for the listings by Greg Stein · 2 days ago
  19. 9fbee0e reconcile: the deployment key, frozen with tests by Greg Stein · 2 days ago
  20. 62c00db docs: fleet reconcile, the not-reached grace clock and the listen URL fallback by Greg Stein · 2 days ago
  21. 22ac061 docs: one absence rule for both GPU listings by Greg Stein · 2 days ago
  22. 158da5c docs: fleet reconcile, unlisted from the listings alone, and early work by Greg Stein · 2 days ago
  23. 79238b7 docs: pin reconcile passes, and hold the unlisted sentence by Greg Stein · 2 days ago
  24. 5e33382 docs: fleet reconcile, the review folded in by Greg Stein · 2 days ago
  25. 03d5ba2 docs: pin the fleet reconcile gates by Greg Stein · 2 days ago
  26. 0fa0a89 docs: expand the fleet reconcile draft by Greg Stein · 3 days ago
  27. 97d2ee5 docs: fleet reconciliation design, a draft by Greg Stein · 3 days ago
  28. 131e954 bin: suppress N813 on the EasyDict import, as the rest of the tree does by Andrew Musselman · 3 days ago
  29. 65e22c8 fleet: a serving box keeps its route, and a stale one is removed by Andrew Musselman · 3 days ago
  30. 676dba9 docs: access control, corrections from reading LiteLLM 1.105.0 by Andrew Musselman · 3 days ago
  31. 8a8aab8 Added a script to validate config before run time by Andrew Musselman · 3 days ago
  32. 1fa0d98 Merge pull request #51 from apache/gs/claude-doc by Andrew Musselman · 3 days ago
  33. b32a90b docs: access control design, a draft by Greg Stein · 3 days ago
  34. e1567cb fleet: drop a gone box's LiteLLM deployment by itself, and offer no buttons by Greg Stein · 3 days ago
  35. 243fff1 fleet: say when no provider lists a host, and ask RunPod on its own schedule by Greg Stein · 3 days ago
  36. 37c9a91 Shorter None guards where the long form says nothing more by Greg Stein · 3 days ago
  37. a2cad72 models: say where each deployment lives by Greg Stein · 3 days ago
  38. 2319ad6 models: say what each model is good for, and not where it runs by Greg Stein · 3 days ago
  39. 7b39df1 capacity: the /admin/capacity page by Greg Stein · 4 days ago
  40. 640a799 capacity: what the model providers we pay were paid by Greg Stein · 4 days ago
  41. 0c994a7 capacity: each hour carries its server's share of the box's rental by Greg Stein · 4 days ago
  42. d13342e sampler: the one reader of /metrics, KV cache size included by Greg Stein · 4 days ago
  43. 99abf5e admin: one page that says what is waiting, and what admins have done by Greg Stein · 4 days ago
  44. d52192d admin: a short log of admin actions, and who last changed each tier by Greg Stein · 4 days ago
  45. fcc038c pruner: a daily runner for the memory store, starting with requests by Greg Stein · 4 days ago
  46. 759a6e3 sampler: one reader of the boxes' /metrics, feeding the hourly rollup by Greg Stein · 5 days ago
  47. 0135d40 capacity: hourly rollup of vLLM readings, kept in LiteLLM's memory store by Greg Stein · 5 days ago
  48. 2c0c252 models: drop the derived per-token pricing for self-hosted models by Greg Stein · 5 days ago
  49. f943a62 config.yaml.example: one comment for providers, site_admins' back on it by Greg Stein · 5 days ago
  50. 300e940 providers: the Vast and RunPod keys are provider accounts by Greg Stein · 5 days ago
  51. ef6532c startup: an ERROR when LiteLLM never refreshes its price list by Greg Stein · 5 days ago
  52. ba0a079 models: say why a serving deployment's badge is yellow by Greg Stein · 7 days ago
  53. ea582e7 install_set: run each server with vLLM's startup plan by Greg Stein · 7 days ago
  54. 8e09549 install_set: keep at least 2GB of each card outside vLLM's budget by Greg Stein · 7 days ago
  55. 105bec2 install_set: use every GPU, and fit the KV cache as well as the weights by Greg Stein · 7 days ago
  56. c6c2041 skew: a deployment at the listen port is reported, not counted by Greg Stein · 7 days ago
  57. a71393d deployments: an unknown address is unknown, not missing by Greg Stein · 7 days ago
  58. bc1d40a fleet: compare with LiteLLM only after the fleet's first pass by Greg Stein · 7 days ago
  59. be13d43 hosting/vast: how to re-provision a box by Greg Stein · 7 days ago
  60. fbca6bc install_set: a re-provision replaces the servers already running by Greg Stein · 7 days ago
  61. 8426d43 fleet: a box serves its model under the name LiteLLM asks for by Greg Stein · 7 days ago
  62. 87fb0de deployments: one failing deployment no longer stops the rest by Greg Stein · 7 days ago
  63. 81ce70c providers: a deployment says who serves a model, not models.yaml by Greg Stein · 7 days ago
  64. 15158b9 hosting/runpod: a pod names itself by its pod id by Greg Stein · 7 days ago
  65. c42e2ec hosting/runpod: how to check the image locally with Podman by Greg Stein · 7 days ago
  66. 7582ece hosting/runpod: build the image on GitHub and push it to ghcr by Greg Stein · 7 days ago
  67. 0920aed hosting/runpod: an image that provisions once, then runs supervisord by Greg Stein · 7 days ago
  68. 6deff9a hosting: one installer for every provider by Greg Stein · 7 days ago
  69. c93881c edict for the dictionaries llmao builds and reads itself by Greg Stein · 8 days ago
  70. b6a9ff9 move the LiteLLM hook beside main.py by Greg Stein · 10 days ago
  71. 323e6a1 models: self_hosted is no longer written in models.yaml by Greg Stein · 10 days ago
  72. 3e3c4ac seam: changing a tier no longer throws away cached usage by Greg Stein · 10 days ago
  73. 23c80fd Making budgeet script +x by Andrew Musselman · 10 days ago
  74. 1a7c5eb bin: measure the max_tokens floor each model needs by Andrew Musselman · 10 days ago
  75. 9274711 proxy: reject an empty completion instead of returning 200 by Andrew Musselman · 10 days ago
  76. 33eaa59 Linting by Andrew Musselman · 10 days ago
  77. 9a6c82f teams: the wrong project is the caller's mistake, not LiteLLM being down by Greg Stein · 10 days ago
  78. 01114f1 proxy: watch for edits made outside llmao, and show them as work needed by Greg Stein · 10 days ago
  79. 8ff46a9 bump LiteLLM to 105 by Greg Stein · 10 days ago
  80. 5d4f3fd pages: three things a page tells a person, kept apart, on every page by Greg Stein · 10 days ago
  81. 4737f43 pages: show skew and log, and let a log entry be acknowledged by Greg Stein · 10 days ago
  82. 13f5881 proxy: a place to tell people things -- skew and log by Greg Stein · 10 days ago
  83. b690d48 projects: a key with no project may use only our own boxes by Greg Stein · 10 days ago
  84. 58bf0cf usage: the allowance counts only what a person's free-tier keys drew by Greg Stein · 10 days ago
  85. 6babc91 tests: placeholder user ids in place of real ones by Greg Stein · 10 days ago
  86. d4b6284 projects: make the pretend projects' teams at startup by Greg Stein · 10 days ago
  87. 02314ae projects: two pretend projects for keys that belong to none by Greg Stein · 10 days ago
  88. 94ec5e4 proxy: fix TeamManager.info for a project with no team yet by Greg Stein · 10 days ago
  89. e3b27cc tests: split what was left of test_entitlements by what it tests by Greg Stein · 11 days ago
  90. 18fe212 tests: the tier tests get a file of their own by Greg Stein · 11 days ago
  91. 21bc158 tests: TeamManager's tests get a file of their own by Greg Stein · 11 days ago
  92. 895c82a tests: name two test files for what they test. No functional change by Greg Stein · 11 days ago
  93. 8a5af10 tests: test_phase1 and test_projects become test_seam by Greg Stein · 11 days ago
  94. 33724d8 tiers: mint keys on the right tier, with the tier's current limits by Greg Stein · 11 days ago
  95. 8a8ede1 proxy: LiteLLMBackend keeps only what it uses by Greg Stein · 11 days ago
  96. 196bad8 proxy: remove the last pass-throughs; the seam talks to the managers by Greg Stein · 11 days ago
  97. 1f5a8eb proxy: delete the Backend protocol and the pass-throughs only it needed by Greg Stein · 11 days ago
  98. cd76f08 proxy: KeyManager, the last class cut from LiteLLMBackend by Greg Stein · 11 days ago
  99. 6c59ecc proxy: UsageManager, cut from the end of LiteLLMBackend by Greg Stein · 11 days ago
  100. 93daeb7 proxy: TeamManager, cut from the end of LiteLLMBackend by Greg Stein · 11 days ago