Skip to content

Conversation

@franciscojavierarceo
Copy link
Collaborator

@franciscojavierarceo franciscojavierarceo commented Oct 17, 2025

What does this PR do?

Migrate vector-io integration tests from --stack-config to use the starter distro and llama stack build, which is a much bigger reliability benefit.

Requires filtering vector-io providers for integration tests.

Also fixed a type issue from post training that was exposed from doing this.

Splitting this out from #3818

Test Plan

Added unit tests and CI

@meta-cla meta-cla bot added the CLA Signed This label is managed by the Meta Open Source bot. label Oct 17, 2025
@cdoern
Copy link
Contributor

cdoern commented Oct 17, 2025

fyi, this might get nuked by #2974 where I already fixed this up

@franciscojavierarceo franciscojavierarceo force-pushed the update-vector-io-tests branch 2 times, most recently from 74debcb to 8d99350 Compare October 17, 2025 14:43
- name: Build Llama Stack
run: |
uv run --no-sync llama stack build --template ci-tests --image-type venv
uv run --no-sync llama stack build --distro starter --image-type venv --single-provider "vector_io=${{ matrix.vector-io-provider }}"
Copy link
Contributor

Choose a reason for hiding this comment

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

curious, should we stick with ci-tests distro similar to other integration tests? Or is there a benefit to starter here

)

self.parser.add_argument(
"--single-provider",
Copy link
Contributor

Choose a reason for hiding this comment

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

nice. I will migrate this to list-deps too

Signed-off-by: Francisco Javier Arceo <[email protected]>
@ashwinb
Copy link
Contributor

ashwinb commented Oct 17, 2025

@franciscojavierarceo Why do we need to add a --single-provider option? I think it is much simpler if you do the following:

-- pass the --stack-config a=b,c=d line locally
-- look at the llama_stack_client fixture which generates adhoc config, save it
-- then paste in the run.yaml into the github workflow

and that's it. NO OTHER CHANGES needed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed This label is managed by the Meta Open Source bot.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants