v0.2.4

Highlights

• One-liner to install and run Llama Stack yay! by @reluctantfuturist in #1383
• support for NVIDIA NeMo datastore by @raspawar in #1852
• (yuge!) Kubernetes authentication by @leseb in #1778
• (yuge!) OpenAI Responses API by @bbrowning in #1989
• add api.llama provider, llama-guard-4 model by @ashwinb in #2058

What's Changed

• docs: update prompt_format.md for llama4 by @ehhuang in #2035
• fix: updated watsonx inference chat apis with new repo changes by @Sajikumarjs in #2033
• fix: Bump h11 to 0.16.0 to fix cve-2025-43859 by @terrytangyuan in #2041
• docs: Add changelog for v0.2.2 and v0.2.3 by @terrytangyuan in #2040
• feat: Llama Stack Meta Reference installation script by @reluctantfuturist in #1383
• chore(github-deps): bump actions/setup-python from 5.5.0 to 5.6.0 by @dependabot in #2038
• feat: Add NVIDIA NeMo datastore by @raspawar in #1852
• feat: Add Kubernetes authentication by @leseb in #1778
• feat: OpenAI Responses API by @bbrowning in #1989
• ci: simplify external provider integration test by @leseb in #2050
• fix: tools page on playground resets agent after every interaction by @MichaelClifford in #2044
• fix: add todo for schema validation by @KPostOffice in #1991
• fix: ollama still using tools with tool_choice="none" by @bbrowning in #2047
• feat: add api.llama provider, llama-guard-4 model by @ashwinb in #2058

Full Changelog: v0.2.3...v0.2.4