Commit graph

  • ec4d0ade09
    chore(deps): bump aiohttp from 3.12.13 to 3.12.14 dependabot[bot] 2025-07-24 23:21:17 +00:00
  • e669e03e87
    chore(deps): bump form-data from 4.0.2 to 4.0.4 in /llama_stack/ui dependabot[bot] 2025-07-24 23:21:00 +00:00
  • 33fabcf1fc
    chore(deps): bump starlette from 0.45.3 to 0.47.2 dependabot[bot] 2025-07-24 23:20:57 +00:00
  • ebea3c8277 api access Eric Huang 2025-07-24 14:56:17 -07:00
  • 2defebc835 chore: Enabling teste for Weaviate Francisco Javier Arceo 2025-07-23 21:20:16 -04:00
  • 110ed98e02
    Update external.py Ashwin Bharambe 2025-07-24 13:33:54 -07:00
  • b7792ec1ee
    Update external.py (minor) Ashwin Bharambe 2025-07-24 13:32:08 -07:00
  • 3ddf8c1b84
    feat: disable post training in starter but keep the CI test Sébastien Han 2025-07-24 22:09:27 +02:00
  • 49fa26949a
    fix: lazy import training recipe Charlie Doern 2025-07-22 16:05:38 -04:00
  • 8125743212
    chore: revert "feat(ci): add a ci-tests distro (#2826)" Sébastien Han 2025-07-21 09:48:03 +02:00
  • 67b6f79715
    ci: tests Sébastien Han 2025-06-12 16:39:08 +02:00
  • e258290213 fix: use logger for console telemetry Charlie Doern 2025-07-21 15:55:43 -04:00
  • d34f7f833f docs: Update nvidia docs template Kelly Brown 2025-07-24 15:52:25 -04:00
  • 5fdd4952a9
    feat: Bring Your Own API (BYOA) Sébastien Han 2025-05-22 17:40:45 +02:00
  • 5e98abcb9a
    chore: install script should use starter Sébastien Han 2025-07-24 21:07:13 +02:00
  • e0206795d9 refactor: enhance route system with WebMethod metadata Eric Huang 2025-07-24 10:49:50 -07:00
  • 0fe110d94a make refreshing happen for all routing tables, naming changes, ollama fixes Ashwin Bharambe 2025-07-24 10:24:10 -07:00
  • 48c5d089c6 Merge remote-tracking branch 'upstream/main' into update-api-docs Sai Soundararaj 2025-07-24 09:53:57 -07:00
  • 63df2886ec
    fix: various improvements on install.sh Sébastien Han 2025-07-11 11:13:11 +02:00
  • 7420c1db11 test: Add VLLM provider support to integration tests Derek Higgins 2025-07-14 11:38:10 +01:00
  • 3e7ea4dd14 Fix unit test client calls Derek Higgins 2025-07-11 09:35:56 +01:00
  • 561912064c
    Merge branch 'main' into fix/issue-2584-llama4-tool-calling-v2 Sumanth Kamenani 2025-07-24 09:18:34 -04:00
  • 617f171923 fix: starter template and litellm backward compat conflict for openai Matthew Farrellee 2025-07-24 08:28:06 -04:00
  • 0de8edd71d feat: switch to Python-based container build system Derek Higgins 2025-07-24 11:02:51 +01:00
  • b04d92ed72 refactor: clean up generated Python Derek Higgins 2025-07-24 10:01:46 +01:00
  • 58435f7579 feat: convert build_container.sh to Python Derek Higgins 2025-07-24 09:16:43 +01:00
  • 64e08e372b chore(test): migrate unit tests from unittest to pytest nvidia test safety Mustafa Elbehery 2025-07-17 12:15:03 +02:00
  • 487e073378 library client fix since we need a runloop for stack construction which can create forever running background threads Ashwin Bharambe 2025-07-23 18:16:16 -07:00
  • 3cda82be3a fix import Ashwin Bharambe 2025-07-23 16:36:24 -07:00
  • 8fb4feeba1 add support embedding models and keeping provider models separate Ashwin Bharambe 2025-07-23 16:13:47 -07:00
  • cf629f81fe cancel refresh task on shutdown Ashwin Bharambe 2025-07-22 16:20:52 -07:00
  • e3396513e9 add configuration to control which models are exposed Ashwin Bharambe 2025-07-22 16:16:08 -07:00
  • 2e5ffab4e3 feat(registry): make the Stack query providers for model listing Ashwin Bharambe 2025-07-22 14:13:21 -07:00
  • a9cfc7df95
    Merge branch 'main' into fix/issue-2584-llama4-tool-calling-v2 Sumanth Kamenani 2025-07-23 16:39:07 -04:00
  • befb2961c0
    Merge branch 'main' into chromadb_openai_compatible Francisco Arceo 2025-07-23 16:13:59 -04:00
  • 305b1bc735 exposing files api tests Francisco Javier Arceo 2025-07-23 16:10:55 -04:00
  • 67307a8949 updating starter to include kv store path and update unit test packages to include chromadb inline Francisco Javier Arceo 2025-07-23 16:04:41 -04:00
  • f5c1935c18 fix: Resolve Llama4 tool calling 500 errors skamenan7 2025-07-21 14:12:55 -04:00
  • 3d43e143d2 refactor: make sku_list resolve provider aliases generically skamenan7 2025-07-21 14:12:24 -04:00
  • 56d01cee69
    Merge branch 'main' into chromadb_openai_compatible Francisco Arceo 2025-07-23 15:15:49 -04:00
  • 0c24d0cc41 updated tests and adpaters to include chroma Francisco Javier Arceo 2025-07-23 15:14:31 -04:00
  • 6fc8b7a93d fix: bring back dell template Ashwin Bharambe 2025-07-23 11:36:05 -07:00
  • 41a45580e0 added a comment in the tests & re run tests upstream, works locally Ubuntu 2025-07-23 18:22:30 +00:00
  • 736404c1bd removed more redunant code Ubuntu 2025-07-23 18:05:29 +00:00
  • 6cd339a2f2 chore: Added openai compatible vector io endpoints for chromadb sarthakdeshpande 2025-06-22 22:05:20 +05:30
  • 41f4678faf with util functions and working Ubuntu 2025-07-23 17:47:34 +00:00
  • 2adc228762
    feat: create dynamic model registration for Anthropic remote inference provider r3v5 2025-07-23 18:45:46 +01:00
  • 7ca4418344 back to working Ubuntu 2025-07-23 17:38:58 +00:00
  • a729f6575f
    fix: fixed test_access_control.py unit test r3v5 2025-07-23 17:46:09 +01:00
  • c871c9d0ac fix: update check-workflows-use-hashes to use github error format Mohit Gaur 2025-07-23 16:24:39 +00:00
  • 10ba79d352
    docs: Update CHANGELOG.md Yuan Tang 2025-07-23 11:46:09 -04:00
  • 1c7be17113 feat: enable DPO training with HuggingFace inline provider Ubuntu 2025-07-23 15:39:36 +00:00
  • 07ae065aeb
    Merge branch 'main' into migrate-vector-store-helpers Francisco Arceo 2025-07-23 09:57:33 -04:00
  • ecbb336e0a fix: prevent shell redirection issues with pip dependencies Derek Higgins 2025-07-23 10:00:17 +01:00
  • 167a257fbd docs: Document use cases for Responses and Agents APIs ChristianZaccaria 2025-07-15 11:12:26 +01:00
  • 5234be70d5 fix: cleanup after build_container.sh Derek Higgins 2025-07-23 11:55:35 +01:00
  • 3e2a9f329e chore: Moving vector store and vector store files helper methods to openai_vector_store_mixin Francisco Javier Arceo 2025-07-22 23:56:28 -04:00
  • be2691212b fix Ashwin Bharambe 2025-07-22 15:10:08 -07:00
  • 4754f6dd95 add warning Ashwin Bharambe 2025-07-22 14:29:16 -07:00
  • 407c3e3bad feat: use XDG directory standards Mustafa Elbehery 2025-07-03 18:48:53 +02:00
  • 447e359f97 add OpenAIMixin to the docs for new api provider authors Matthew Farrellee 2025-07-22 16:58:24 -04:00
  • abdfc47017
    fix: honour deprecation of --config and --template Sébastien Han 2025-07-22 15:37:55 +02:00
  • 6e17e2ccf2 Fix warnings in builds for documentation Kelly Brown 2025-07-16 11:37:38 -04:00
  • 81361add26 chore(test): fix flaky telemetry tests Mustafa Elbehery 2025-07-18 18:38:37 +02:00
  • d76b3aa4d2 more fixes Ashwin Bharambe 2025-07-22 10:45:39 -07:00
  • a66074a10e kill older tech-debt, make get_provider_impl async Ashwin Bharambe 2025-07-22 10:20:04 -07:00
  • 38a9c119df must override get_provider_impl also Ashwin Bharambe 2025-07-22 10:13:27 -07:00
  • 50d16dc707 refactor lookup_model out so it can be used by vector dbs routing table Ashwin Bharambe 2025-07-22 10:08:29 -07:00
  • 91f169ab77 fix: optimize container build by enabling uv cache Derek Higgins 2025-07-22 12:46:42 +01:00
  • d3dee496ec test: add comprehensive tests for enhanced model registration and lookup Ashwin Bharambe 2025-07-22 09:21:17 -07:00
  • 352bf3ec56 feat(registry): more flexible model lookup Ashwin Bharambe 2025-07-22 09:08:51 -07:00
  • ac9bd02915
    test: modify mode validation tests to be more robust Bobbins228 2025-07-22 16:55:40 +01:00
  • 3d9f83ae87
    fix: search mode validation for rag query Bobbins228 2025-07-22 16:36:21 +01:00
  • 4c08337b9e chore: Making name optional in openai_create_vector_store Francisco Javier Arceo 2025-07-22 12:08:04 -04:00
  • 20d401e953 use llama_stack.log.get_logger Matthew Farrellee 2025-07-22 10:30:59 -04:00
  • 0ff9ae01a0 Merge branch 'main' into openai-mixin Matthew Farrellee 2025-07-22 10:27:10 -04:00
  • 02e13617c2
    Merge branch 'main' into feat/2729-configurable-embeddings-v2 Sumanth Kamenani 2025-07-22 08:50:01 -04:00
  • a39848e51b fix(agent): ensure turns are sorted Omer Tuchfeld 2025-07-22 13:56:20 +02:00
  • a5309d6ff1 fix(install): explicit docker.io usage Omer Tuchfeld 2025-07-22 10:55:09 +02:00
  • 99dace6c78 chore: remove *_openai_compat providers Eric Huang 2025-07-21 22:22:03 -07:00
  • 6411b053ec
    Merge branch 'main' into load-quickstart-py Francisco Arceo 2025-07-21 22:51:49 -04:00
  • a38770edab chore: Adding demo script and importing it into the docs Francisco Javier Arceo 2025-07-21 22:45:19 -04:00
  • 29dc5c44dd updated pgvector description Jeremy Choi 2025-07-22 09:56:44 +10:00
  • ebf12d0ab0 chore(tests): replace unicode punctuation in configurable embeddings tests skamenan7 2025-07-17 13:45:50 -04:00
  • d55dd3e9a0 feat(vector-io): configurable embedding models for all providers (v2)\n\nAdds embedding_model and embedding_dimension fields to all VectorIOConfig classes.\nRouter respects provider defaults with fallback.\nIntroduces embedding_utils helper.\nComprehensive docs & samples.\nResolves #2729 skamenan7 2025-07-17 11:51:40 -04:00
  • b5aa8c7aa2
    Merge branch 'main' into acl-vector-stores Francisco Arceo 2025-07-21 15:51:37 -04:00
  • d58c2c5f3c renaming authorize_action to assert_action_allowed Francisco Javier Arceo 2025-07-21 15:47:43 -04:00
  • 9a308e0ef1
    Merge branch 'main' into acl-vector-stores Francisco Arceo 2025-07-21 15:27:51 -04:00
  • dffacae2de fix: uvicorn respect log_config Charlie Doern 2025-07-21 15:03:13 -04:00
  • 20ea53308a unify server, run, simplify utils Eric Huang 2025-07-21 11:42:42 -07:00
  • a859bf47ee
    Merge branch 'main' into acl-vector-stores Francisco Arceo 2025-07-21 13:51:52 -04:00
  • ae4eab3bd2
    Merge branch 'main' into acl-vector-stores Francisco Arceo 2025-07-21 12:24:30 -04:00
  • 066d7705db removing @pytest.mark.asyncio or @pytest_asyncio.fixture Francisco Javier Arceo 2025-07-21 11:24:00 -04:00
  • c248b8c69a
    Merge branch 'main' into acl-vector-stores Francisco Arceo 2025-07-21 11:12:34 -04:00
  • e60c0f623f
    fix: remove @pytest.mark.asyncio from test_get_raw_document_text.py r3v5 2025-07-21 16:12:32 +01:00
  • 639bc912d5 chore: create OpenAIMixin for inference providers with an OpenAI-compat API that need to implement openai_* methods Matthew Farrellee 2025-07-21 07:27:27 -04:00
  • 1c9b7446a0 fix(vectordb): VectorDBInput has no provider_id Mustafa Elbehery 2025-07-21 10:24:25 +02:00
  • cbeab1af22 Add permissions for pull request creation in coverage-badge workflow ChristianZaccaria 2025-07-21 10:34:01 +01:00
  • c67bae2d07 Merge branch 'main' into allow-dynamic-models-ollama Matthew Farrellee 2025-07-21 05:17:29 -04:00
  • cb0321a034
    fix: graceful SIGINT on server Sébastien Han 2025-07-21 11:16:56 +02:00