Llama Stack Files

Highlights

• feat(tests): introduce inference record/replay to increase test reliability by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2941
• fix(library_client): improve initialization error handling and prevent AttributeError by @mattf in https://github.com/meta-llama/llama-stack/pull/2944
• fix: use OLLAMA_URL to activate Ollama provider in starter by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2963
• feat(UI): adding MVP playground UI by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2828
• Standardization of errors (@nathan-weinberg)
• feat: Enable DPO training with HuggingFace inline provider by @Nehanth in https://github.com/meta-llama/llama-stack/pull/2825
• chore: rename templates to distributions by @ashwinb in https://github.com/meta-llama/llama-stack/pull/3035

What's Changed

• docs: update using llama stack as library docs by @Bobbins228 in https://github.com/meta-llama/llama-stack/pull/2931
• feat: add base64 encoded PDF support for OpenAI Chat Completions by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2881
• fix: error on failed job, do not wait for timeout by @mattf in https://github.com/meta-llama/llama-stack/pull/2945
• fix: Update SFTConfig parameter to fix CI and Post Training Workflow by @Nehanth in https://github.com/meta-llama/llama-stack/pull/2948
• feat(tests): introduce inference record/replay to increase test reliability by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2941
• chore: Delete coverage-badge by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2950
• fix: use same image_name logic for build & run config by @mattf in https://github.com/meta-llama/llama-stack/pull/2949
• feat(test): record agents, safety and vector_io integration tests by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2952
• chore(ci): disable post training tests by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2953
• feat(tests): record responses for evals and telemetry tests by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2954
• feat(ci): use replay mode, setup ollama if specific label exists on PR by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2955
• feat(ci): keep only one re-recording job because independent recordings will conflict by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2956
• fix(library_client): improve initialization error handling and prevent AttributeError by @mattf in https://github.com/meta-llama/llama-stack/pull/2944
• fix: use OLLAMA_URL to activate Ollama provider in starter by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2963
• chore: fix k8s config by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2959
• docs: part 1 - fix warnings in documentation generation by @kelbrown20 in https://github.com/meta-llama/llama-stack/pull/2861
• fix(ci): slightly update workflow trigger by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2966
• chore: standardize model not found error by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2964
• fix: fix No module named 'ollama' in test_inference_recordings.py by @r3v5 in https://github.com/meta-llama/llama-stack/pull/2967
• chore: standardize dataset not found error by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2962
• chore: standardize vector store not found error by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2968
• docs: Add detailed docstrings to API models and update OpenAPI spec by @saiprashanths in https://github.com/meta-llama/llama-stack/pull/2889
• fix: remove redundant code from unregister_dataset by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2971
• fix(ci): more correct concurrency key for workflows by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2973
• feat(UI): adding MVP playground UI by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2828
• chore(rename): move llama_stack.distribution to llama_stack.core by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2975
• feat: Enable DPO training with HuggingFace inline provider by @Nehanth in https://github.com/meta-llama/llama-stack/pull/2825
• fix: post_training ci by @cdoern in https://github.com/meta-llama/llama-stack/pull/2984
• fix: remove unused DPO parameters from schema and tests by @Nehanth in https://github.com/meta-llama/llama-stack/pull/2988
• docs: Reformat external provider documentation by @kelbrown20 in https://github.com/meta-llama/llama-stack/pull/2982
• chore: Remove coverage badge from README.md by @ChristianZaccaria in https://github.com/meta-llama/llama-stack/pull/2976
• fix: remove redundant code from unregister_vector_db by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2983
• fix: Post Training Model change in Tests in order to make it less intensive by @Nehanth in https://github.com/meta-llama/llama-stack/pull/2991
• fix: integration tests not triggering on PR open by @cdoern in https://github.com/meta-llama/llama-stack/pull/2985
• feat(ci): add support for running vision inference tests by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2972
• feat: Add clear error message when API key is missing by @mattf in https://github.com/meta-llama/llama-stack/pull/2992
• fix(ci): simplify integration tests replay mode by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2997
• feat(ci): only run on 3.12, run on both 3.12 and 3.13 nightly by @ashwinb in https://github.com/meta-llama/llama-stack/pull/3000
• chore: Enabling Integration tests for Weaviate by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2882
• feat(ci): introduce workflow for re-recording inference outputs by @ashwinb in https://github.com/meta-llama/llama-stack/pull/3002
• test: use llama stack build when starting server by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2999
• feat: Add openAI compatible APIs to Qdrant by @varshaprasad96 in https://github.com/meta-llama/llama-stack/pull/2465
• chore: Update README for supported DBs by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/3005
• fix: sambanova inference provider by @mattf in https://github.com/meta-llama/llama-stack/pull/2996
• fix(recording): endpoint resolution by @ehhuang in https://github.com/meta-llama/llama-stack/pull/3013
• chore: create integration-tests script by @ehhuang in https://github.com/meta-llama/llama-stack/pull/3016
• refactor: remove Conda support from Llama Stack by @r3v5 in https://github.com/meta-llama/llama-stack/pull/2969
• chore(python-deps): bump huggingface-hub from 0.34.2 to 0.34.3 by @dependabot[bot] in https://github.com/meta-llama/llama-stack/pull/3028
• chore: CI action names by @ehhuang in https://github.com/meta-llama/llama-stack/pull/3014
• refactor: Remove double filtering based on score threshold by @varshaprasad96 in https://github.com/meta-llama/llama-stack/pull/3019
• test: Implement vector store search test by @varshaprasad96 in https://github.com/meta-llama/llama-stack/pull/3001
• feat: switch to async completion in LiteLLM OpenAI mixin by @eranco74 in https://github.com/meta-llama/llama-stack/pull/3029
• chore(ci): remove reportlab dep by @mattf in https://github.com/meta-llama/llama-stack/pull/3033
• chore(python-deps): bump weaviate-client from 4.16.4 to 4.16.5 by @dependabot[bot] in https://github.com/meta-llama/llama-stack/pull/3026
• chore(python-deps): bump opentelemetry-exporter-otlp-proto-http from 1.35.0 to 1.36.0 by @dependabot[bot] in https://github.com/meta-llama/llama-stack/pull/3027
• chore: rename templates to distributions by @ashwinb in https://github.com/meta-llama/llama-stack/pull/3035
• chore(python-deps): bump openai from 1.97.1 to 1.98.0 by @dependabot[bot] in https://github.com/meta-llama/llama-stack/pull/3025
• chore: standardize tool group not found error by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2986
• chore: standardize session not found error by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/3031
• chore: Fixing Markdown renderer by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/3038
• chore: fix: integration tests failures marked as successful by @ehhuang in https://github.com/meta-llama/llama-stack/pull/3039

New Contributors

• @saiprashanths made their first contribution in https://github.com/meta-llama/llama-stack/pull/2889

Full Changelog: https://github.com/meta-llama/llama-stack/compare/v0.2.16...v0.2.17