Llama Stack Files

What's Changed

• fix: handle encoding errors when adding files to vector store by @derekhiggins in https://github.com/meta-llama/llama-stack/pull/2574
• feat: consolidate most distros into "starter" by @leseb in https://github.com/meta-llama/llama-stack/pull/2516
• fix: only load mcp when enabled in tool_group by @leseb in https://github.com/meta-llama/llama-stack/pull/2621
• fix(starter): Add missing faiss provider to build.yaml vector_io section by @derekhiggins in https://github.com/meta-llama/llama-stack/pull/2625
• feat: improve telemetry by @leseb in https://github.com/meta-llama/llama-stack/pull/2590
• docs: Update starter docs to include milvus inline by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2631
• docs: update docs to use "starter" than "ollama" by @zdtsw in https://github.com/meta-llama/llama-stack/pull/2629
• chore(deps): bump next from 15.3.2 to 15.3.3 in /llama_stack/ui by @dependabot[bot] in https://github.com/meta-llama/llama-stack/pull/2632
• refactor: set proper name for embedding all-minilm:l6-v2 and update to use "starter" in detailed_tutorial by @zdtsw in https://github.com/meta-llama/llama-stack/pull/2627
• ci: error when a pipefails by @leseb in https://github.com/meta-llama/llama-stack/pull/2635
• fix: print proper template path upon build by @cdoern in https://github.com/meta-llama/llama-stack/pull/2642
• fix: Restore the nvidia distro by @bbrowning in https://github.com/meta-llama/llama-stack/pull/2639
• fix: authorized sql store with postgres by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2641
• fix: use --template flag for server by @cdoern in https://github.com/meta-llama/llama-stack/pull/2643
• chore: Adding unit tests for Milvus and OpenAI compatibility by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2640
• feat(auth): support github tokens by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2509
• feat(auth,ui): support github sign-in in the UI by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2545
• fix: update k8s templates by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2645
• chore(api): add mypy coverage to prompts by @Elbehery in https://github.com/meta-llama/llama-stack/pull/2657
• fix(remote:milvus): add missing files_api parameter and kvstore configuration by @pgustafs in https://github.com/meta-llama/llama-stack/pull/2630
• chore(api): add mypy coverage to meta_reference_safety by @Elbehery in https://github.com/meta-llama/llama-stack/pull/2661
• chore(api): add mypy coverage to meta_reference_config by @Elbehery in https://github.com/meta-llama/llama-stack/pull/2664
• chore: add mypy loader by @Elbehery in https://github.com/meta-llama/llama-stack/pull/2672
• fix: runpod transition to python 3.12 by @leseb in https://github.com/meta-llama/llama-stack/pull/2682
• chore(api): add mypy coverage to apis by @Elbehery in https://github.com/meta-llama/llama-stack/pull/2648
• docs: Update links to Android Demo App by @jgarciao in https://github.com/meta-llama/llama-stack/pull/2687
• chore: add mypy post training by @Elbehery in https://github.com/meta-llama/llama-stack/pull/2675
• ci: test safety with starter by @leseb in https://github.com/meta-llama/llama-stack/pull/2628
• chore: Update CODEOWNERS by @mattf in https://github.com/meta-llama/llama-stack/pull/2692
• build: replace "python-jose" with "python-jose[cryptography]" by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2695
• fix: llama stack run starter in conda by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2679
• test: terminate server process when finished by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2700
• ci: run all APIs integration tests by @leseb in https://github.com/meta-llama/llama-stack/pull/2646
• fix: properly represent paths in server logs by @cdoern in https://github.com/meta-llama/llama-stack/pull/2698
• chore: update pre-commit hook versions by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2708
• chore(api): add mypy coverage to cli/stack by @Elbehery in https://github.com/meta-llama/llama-stack/pull/2650
• ci: add config for pre-commit.ci by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2712
• chore: Adding unit tests for OpenAI vector stores, making the unit tests extensible for other providers, and migrating SQLite-vec registry to kvstore by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2665
• docs: fix a few broken things in the CONTRIBUTING.md by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2714
• chore: move "install.sh" script into "scripts" dir by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2719
• chore: remove "rfc" directory and move original rfc to "docs" by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2718
• chore: update vllm k8s command to support tool calling by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2717
• fix: auth sql store: user is owner policy by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2674
• fix: container build on podman by @leseb in https://github.com/meta-llama/llama-stack/pull/2723
• fix: pin opentelemtry version by @leseb in https://github.com/meta-llama/llama-stack/pull/2722
• chore: default to pytest asyncio-mode=auto by @mattf in https://github.com/meta-llama/llama-stack/pull/2730
• fix: sambanova shields and model validation by @jhpiedrahitao in https://github.com/meta-llama/llama-stack/pull/2693
• fix: Don't cache clients for passthrough auth providers by @bbrowning in https://github.com/meta-llama/llama-stack/pull/2728
• chore(github-deps): bump medyagh/setup-minikube from 0.0.19 to 0.0.20 by @dependabot[bot] in https://github.com/meta-llama/llama-stack/pull/2738
• chore: block network access from unit tests by @mattf in https://github.com/meta-llama/llama-stack/pull/2732
• feat: Add Vector stores UI by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2737
• feat: add input validation for search mode of rag query config by @Bobbins228 in https://github.com/meta-llama/llama-stack/pull/2275
• docs: clarify run.yaml files are starting points for customization by @skamenan7 in https://github.com/meta-llama/llama-stack/pull/2746
• fix(faiss): Delete file contents from kvstore by @derekhiggins in https://github.com/meta-llama/llama-stack/pull/2686
• feat: add infrastructure to allow inference model discovery by @mattf in https://github.com/meta-llama/llama-stack/pull/2710
• docs: fix building distro link by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2750
• chore: add issue template for technical debt by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2753
• fix: Safety in starter by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2731
• chore: Move vector store kvstore implementation into openai_vector_store_mixin.py by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2748
• feat: Implement keyword search in milvus by @varshaprasad96 in https://github.com/meta-llama/llama-stack/pull/2231
• chore: remove tests.yaml by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2754
• fix: Fix /vector-stores/create API when vector store with duplicate name by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2617
• fix: stop image_name from being cast to an integer by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2759
• docs: update outdated llama stack client documentation by @Bobbins228 in https://github.com/meta-llama/llama-stack/pull/2758
• test: add tests against published client by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2752
• chore: Adding OpenAI Vector Stores Files API compatibility for PGVector by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2755
• docs: Reorganize documentation on the webpage by @kelbrown20 in https://github.com/meta-llama/llama-stack/pull/2651

New Contributors

• @Elbehery made their first contribution in https://github.com/meta-llama/llama-stack/pull/2657
• @pgustafs made their first contribution in https://github.com/meta-llama/llama-stack/pull/2630

Full Changelog: https://github.com/meta-llama/llama-stack/compare/v0.2.14...v0.2.15