Llama Stack Files

Highlights

• search_mode support in OpenAI vector store API
• Security fixes

What's Changed

• fix: stack build by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2485
• chore: upgrade hf hub dependency by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2487
• chore(github-deps): bump astral-sh/setup-uv from 6.0.1 to 6.3.0 by @dependabot in https://github.com/meta-llama/llama-stack/pull/2488
• fix: [#2495] FileNotFound Err in container image by @RohanAwhad in https://github.com/meta-llama/llama-stack/pull/2498
• chore: bump python supported version to 3.12 by @leseb in https://github.com/meta-llama/llama-stack/pull/2475
• fix(ui): ensure initial data fetch only happens once by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2486
• fix: update the signature of openai_list_files_in_vector_store in all VectorIO impls by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2503
• chore: do not build on auth ci test by @leseb in https://github.com/meta-llama/llama-stack/pull/2505
• chore: fix build script bug by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2507
• feat: Add search_mode support to OpenAI vector store API by @varshaprasad96 in https://github.com/meta-llama/llama-stack/pull/2500
• fix: Ollama should be optional in starter distro by @bbrowning in https://github.com/meta-llama/llama-stack/pull/2482
• feat: Add ChunkMetadata to Chunk by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2497
• fix: resume responses with tool call output by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2524
• fix: annotations list and web_search_preview in Responses by @bbrowning in https://github.com/meta-llama/llama-stack/pull/2520
• chore: remove nested imports by @leseb in https://github.com/meta-llama/llama-stack/pull/2515
• fix: finish conversion to StrEnum by @leseb in https://github.com/meta-llama/llama-stack/pull/2514
• refactor(env)!: enhanced environment variable substitution by @leseb in https://github.com/meta-llama/llama-stack/pull/2490
• chore: isolate bare minimum project dependencies by @leseb in https://github.com/meta-llama/llama-stack/pull/2282
• fix: fix test of root span to match what is being set by @grs in https://github.com/meta-llama/llama-stack/pull/2494
• ci: vector_io provider integration tests by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2537
• fix: Some missed env variable changes from PR 2490 by @bbrowning in https://github.com/meta-llama/llama-stack/pull/2538
• docs: Add recent releases to CHANGELOG.md by @terrytangyuan in https://github.com/meta-llama/llama-stack/pull/2533
• build: update temp. created Containerfile by @zdtsw in https://github.com/meta-llama/llama-stack/pull/2492
• fix(security): Upgrade urllib3 to v2.5.0. Fixes CVE-2025-50181 and CVE-2025-50182 by @terrytangyuan in https://github.com/meta-llama/llama-stack/pull/2534
• fix: dataset metadata without provider_id by @JuanmaBM in https://github.com/meta-llama/llama-stack/pull/2527
• fix(security): Upgrade protobuf and aiohttp. Fixes CVE-2025-4565 by @terrytangyuan in https://github.com/meta-llama/llama-stack/pull/2541
• chore: standardize unsupported model error [#2517] by @RohanAwhad in https://github.com/meta-llama/llama-stack/pull/2518
• fix: ValueError in faiss vector database serialization (resolves [#2519]) by @ramkrsna in https://github.com/meta-llama/llama-stack/pull/2526
• fix: proper checkpointing logic for HF trainer by @cdoern in https://github.com/meta-llama/llama-stack/pull/2429
• chore: OpenAI compatibility for Milvus by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2470

New Contributors

• @zdtsw made their first contribution in https://github.com/meta-llama/llama-stack/pull/2492
• @JuanmaBM made their first contribution in https://github.com/meta-llama/llama-stack/pull/2527
• @ramkrsna made their first contribution in https://github.com/meta-llama/llama-stack/pull/2526

Full Changelog: https://github.com/meta-llama/llama-stack/compare/v0.2.12...v0.2.13