Llama Stack Files

Highlights

• OpenAI-compatible vector store APIs
• Hybrid Search in Sqlite-vec
• File search tool in Responses API
• Pagination in inference and response stores
• Added suffix to completions API for fill-in-the-middle tasks

What's Changed

• chore: update CODEOWNERS by @reluctantfuturist in https://github.com/meta-llama/llama-stack/pull/2414
• feat: To add health status check for remote VLLM by @justjais in https://github.com/meta-llama/llama-stack/pull/2303
• fix(faiss): handle case where distance is 0 by setting d to minimum positive… by @Ibrahim-Haroon in https://github.com/meta-llama/llama-stack/pull/2387
• fix(pgvector): handle case where distance is 0 by setting score to infinity by @Ibrahim-Haroon in https://github.com/meta-llama/llama-stack/pull/2416
• fix(security): Upgrade requests to 2.32.4. Fixes CVE-2024-47081 by @terrytangyuan in https://github.com/meta-llama/llama-stack/pull/2425
• docs: Add recent releases by @terrytangyuan in https://github.com/meta-llama/llama-stack/pull/2424
• fix: loosen tool call checks in inference store by @bbrowning in https://github.com/meta-llama/llama-stack/pull/2420
• feat: Add OpenAI compat /v1/vector_store APIs by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2423
• fix: [#1867] InferenceRouter has no attribute formatter by @RohanAwhad in https://github.com/meta-llama/llama-stack/pull/2422
• feat: OpenAIVectorIOMixin for vector_stores common logic by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2427
• fix: set appropriate defaults for params by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2434
• ci: fix external provider test by @leseb in https://github.com/meta-llama/llama-stack/pull/2438
• fix(weaviate): handle case where distance is 0 by setting score to infinity by @Ibrahim-Haroon in https://github.com/meta-llama/llama-stack/pull/2415
• feat: update search for vector_stores by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2441
• feat: update openai tests to work with both clients by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2442
• test: Update integration-tests.yml by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2443
• feat(auth): allow token to be provided for use against jwks endpoint by @grs in https://github.com/meta-llama/llama-stack/pull/2394
• chore: Add OpenAI compatibility for Ollama embeddings by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2440
• feat: File search tool for Responses API by @bbrowning in https://github.com/meta-llama/llama-stack/pull/2426
• feat: Implement hybrid search in SQLite-vec by @varshaprasad96 in https://github.com/meta-llama/llama-stack/pull/2312
• feat: Add suffix to openai_completions by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2449
• feat: Add url field to PaginatedResponse and populate it using route … by @RohanAwhad in https://github.com/meta-llama/llama-stack/pull/2419
• feat: Add Nvidia e2e beginner notebook and tool calling notebook by @JashG in https://github.com/meta-llama/llama-stack/pull/1964
• chore: Add OpenAI compatiblity for vLLM embeddings by @varshaprasad96 in https://github.com/meta-llama/llama-stack/pull/2448
• feat: support pagination in inference/responses stores by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2397
• fix: broken links on nvidia distro docs when rendered by @bbrowning in https://github.com/meta-llama/llama-stack/pull/2446
• fix: clarify bash requirement in install flow by @dltn in https://github.com/meta-llama/llama-stack/pull/2450
• fix: Do not throw when listing vector stores by @hardikjshah in https://github.com/meta-llama/llama-stack/pull/2460

New Contributors

• @justjais made their first contribution in https://github.com/meta-llama/llama-stack/pull/2303
• @Ibrahim-Haroon made their first contribution in https://github.com/meta-llama/llama-stack/pull/2387
• @RohanAwhad made their first contribution in https://github.com/meta-llama/llama-stack/pull/2422

Full Changelog: https://github.com/meta-llama/llama-stack/compare/v0.2.10.1...v0.2.11