Skip to content

[RHIDP-12413] Update for Llama Stack v0.5.x - #7

Merged
Jdubrick merged 4 commits into
redhat-ai-dev:mainfrom
Jdubrick:update-lls-0.5.x
Mar 18, 2026
Merged

[RHIDP-12413] Update for Llama Stack v0.5.x#7
Jdubrick merged 4 commits into
redhat-ai-dev:mainfrom
Jdubrick:update-lls-0.5.x

Conversation

@Jdubrick

@Jdubrick Jdubrick commented Mar 16, 2026

Copy link
Copy Markdown
Contributor

What does this PR do?:

  • Updates the run.yaml to be compatible for Llama Stack v0.5.x
    • Not many changes to be honest
  • Consolidated run.yaml and run-no-guard.yaml, we can use a single file with a ENABLE_SAFETY env var to conditionally add the safety pieces
  • Updated local compose files and commands for local development
  • Updated documentation to reflect new safety handling
  • Updates images.yaml for upcoming Sprint for working versions based on Llama Stack v0.5.x.
    • Our 0.4.3 RAG works with 0.5.x

Which issue(s) this PR fixes:

https://redhat.atlassian.net/browse/RHIDP-12413
https://redhat.atlassian.net/browse/RHIDP-12408

PR acceptance criteria:

Testing and documentation do not need to be complete in order for this PR to be approved. We just need to ensure tracking issues are opened and linked to this PR, if they are not in the PR scope due to various constraints.

  • Tested and Verified

  • Documentation (READMEs, Product Docs, Blogs, Education Modules, etc.)

How to test changes / Special notes to the reviewer:

Signed-off-by: Jordan Dubrick <jdubrick@redhat.com>
Signed-off-by: Jordan Dubrick <jdubrick@redhat.com>
Signed-off-by: Jordan Dubrick <jdubrick@redhat.com>

@JslYoon JslYoon left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there a reason why we keep both WITH_SAFETY and ENABLE_SAFETY for the safety guard configuration?

Comment thread llama-stack-configs/config.yaml
@Jdubrick

Copy link
Copy Markdown
Contributor Author

@JslYoon ENABLE_SAFETY fits our ENABLE_* syntax for conditionally adding items, WITH_SAFETY is just the Makefile syntax if you want the guard up locally. I can standardize it if you prefer

@JslYoon

JslYoon commented Mar 16, 2026

Copy link
Copy Markdown
Contributor

@JslYoon ENABLE_SAFETY fits our ENABLE_* syntax for conditionally adding items, WITH_SAFETY is just the Makefile syntax if you want the guard up locally. I can standardize it if you prefer

Just was confused by two similar names, perhaps standardizing it could be better? I'm open for both

Signed-off-by: Jordan Dubrick <jdubrick@redhat.com>
@Jdubrick

Copy link
Copy Markdown
Contributor Author

@JslYoon I'm inclined to leave them separately named to help make it clear what is for spinning things up locally vs. used in Llama Stack config for filtering out items. If it gets confusing down the line we can re-assess. wdyt?

@Jdubrick
Jdubrick requested review from JslYoon and thepetk March 17, 2026 13:04

@thepetk thepetk left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm @Jdubrick I've tested it with the rolling demo and it worked, apart ofc from the lightspeed frontend part which I don't think is related with this PR

@Jdubrick
Jdubrick merged commit b17d894 into redhat-ai-dev:main Mar 18, 2026
3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants