Skip to content

adds simple smoke tests#76

Open
warmbowski wants to merge 4 commits into
mainfrom
paull/add_labs_smoke_tests_for_deployed_models
Open

adds simple smoke tests#76
warmbowski wants to merge 4 commits into
mainfrom
paull/add_labs_smoke_tests_for_deployed_models

Conversation

@warmbowski

@warmbowski warmbowski commented Feb 27, 2026

Copy link
Copy Markdown
Contributor

Description

Adds some simple smoke tests for running against deployed models on the litellm provider. These are mostly recreate some sanity check logic from this script that Pradeep runs against deployed olmo3 models on deployment: https://github.com/allenai/open-instruct/blob/pd-olmo-api-eval/scripts/evaluate_api_endpoints.py#L125

closes https://github.com/allenai/playground-issues-repo/issues/965

Type of Change

  • Bug fix (non-breaking change that fixes an issue)
  • New feature (non-breaking change that adds functionality)
  • Breaking change (fix or feature that would cause existing functionality to not work as expected)
  • Documentation update
  • Refactoring (no functional changes)

Testing

  • Unit tests pass locally (pytest tests/ --ignore=tests/integration/)
  • Integration tests pass (if applicable)
  • New tests added for new functionality

Checklist

  • My code follows the project's style guidelines
  • I have performed a self-review of my code
  • I have added/updated documentation as needed
  • My changes generate no new warnings
  • Any dependent changes have been merged and published

@warmbowski warmbowski requested a review from undfined February 27, 2026 18:10
@warmbowski warmbowski changed the title adds smoke simple smoke tests adds simple smoke tests Feb 27, 2026
@warmbowski warmbowski force-pushed the paull/add_labs_smoke_tests_for_deployed_models branch from 88eda73 to 4ea4ed0 Compare March 17, 2026 23:58
@warmbowski warmbowski marked this pull request as ready for review March 19, 2026 20:47
extracted_answer: Any = None
metadata: dict[str, Any] = field(default_factory=dict)
tool_calls: list[ToolCall] | None = None
provider_extras: dict[str, Any] = field(default_factory=dict)

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

this is a dict set up for provider specific (in this case litellm) properties from the response. Only thing add in this is has_reasoning. I suppose other providers could also implement this on LMOutput (i.e. embedded thinking tag). Mainly, I am wondering if this is okay to add this dict (can rename, maybe can just be in metadata, but that looks like its for metric calls).



@dataclass(frozen=True, slots=True)
class SubstringScorer(Scorer):

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Look fro existing implementation of this, I think we have something.

@warmbowski warmbowski force-pushed the paull/add_labs_smoke_tests_for_deployed_models branch from 332a01a to e922f33 Compare April 8, 2026 15:51
@warmbowski warmbowski requested a review from undfined April 8, 2026 17:37

@undfined undfined left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

lgtm

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.

2 participants