Llama Stack Files

Highlights

• Support for Llama Guard 4
• Added Milvus support to vector-stores API
• Documentation and zero-to-hero updates for latest APIs

What's Changed

• fix(docs): add setuptools explicitly by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2547
• chore(github-deps): bump astral-sh/setup-uv from 6.3.0 to 6.3.1 by @dependabot in https://github.com/meta-llama/llama-stack/pull/2548
• docs: update document since container is not an option for "llama stack run" + update docs with current "usage" by @zdtsw in https://github.com/meta-llama/llama-stack/pull/2531
• feat: Add webmethod for deleting openai responses by @2000krysztof in https://github.com/meta-llama/llama-stack/pull/2160
• fix: add missing argument and methods by @leseb in https://github.com/meta-llama/llama-stack/pull/2550
• docs: auto generated documentation for providers by @leseb in https://github.com/meta-llama/llama-stack/pull/2543
• fix(ollama): Download remote image URLs for Ollama by @ashwinb in https://github.com/meta-llama/llama-stack/pull/2551
• docs: fixed incorrect API list item by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2566
• fix: default api_key from env must be a SecretStr by @mattf in https://github.com/meta-llama/llama-stack/pull/2565
• fix: update zero-to-hero guide for modern llama stack by @mattf in https://github.com/meta-llama/llama-stack/pull/2555
• fix: Fixing Milvus sample config and updating documentation by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2568
• chore: Enabling VectorIO Integration tests for Milvus by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2546
• docs: update external provider guide and navigation by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2567
• fix: allow default empty vars for conditionals by @leseb in https://github.com/meta-llama/llama-stack/pull/2570
• docs: specify the ability to train non-Llama models by @nathan-weinberg in https://github.com/meta-llama/llama-stack/pull/2573
• fix: update zero_to_hero package and README by @zdtsw in https://github.com/meta-llama/llama-stack/pull/2578
• test: Add one-step integration testing with server auto-start by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2580
• fix: Set parameter usedforsecurity=False when calling hashlib.md5 in order to fix rag_tool.insert on FIPS clusters by @jgarciao in https://github.com/meta-llama/llama-stack/pull/2577
• docs: Minor spelling fix by @nharada1 in https://github.com/meta-llama/llama-stack/pull/2592
• docs: update full list of providers with matched APIs and dockerhub images by @zdtsw in https://github.com/meta-llama/llama-stack/pull/2452
• fix(docs): update Agents101 notebook for builtin websearch by @skamenan7 in https://github.com/meta-llama/llama-stack/pull/2591
• docs: Add quick_start.ipynb notebook equivalent of index.md Quickstart guide by @ChristianZaccaria in https://github.com/meta-llama/llama-stack/pull/2128
• fix: add mcp dependency to agent provider by @leseb in https://github.com/meta-llama/llama-stack/pull/2587
• fix: store configs by @ehhuang in https://github.com/meta-llama/llama-stack/pull/2593
• fix: AccessDeniedError leads to HTTP 500 instead of error 403 by @akram in https://github.com/meta-llama/llama-stack/pull/2595
• ci: let pytest run the distro server by @leseb in https://github.com/meta-llama/llama-stack/pull/2586
• chore: Add support for vector-stores files api for Milvus by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2582
• chore: Updating chunk id generation to ensure uniqueness by @franciscojavierarceo in https://github.com/meta-llama/llama-stack/pull/2618
• fix: CI flakiness in vector IO tests by pinning pymilvus>=2.4.10 by @derekhiggins in https://github.com/meta-llama/llama-stack/pull/2610
• feat: add llama guard 4 model by @mattf in https://github.com/meta-llama/llama-stack/pull/2579

New Contributors

• @nharada1 made their first contribution in https://github.com/meta-llama/llama-stack/pull/2592
• @skamenan7 made their first contribution in https://github.com/meta-llama/llama-stack/pull/2591
• @akram made their first contribution in https://github.com/meta-llama/llama-stack/pull/2595

Full Changelog: https://github.com/meta-llama/llama-stack/compare/v0.2.13...v0.2.14