Skip to content

update mteb script#4265

Merged
pgladkows merged 1 commit into
mainfrom
update-mteb-script
Jun 1, 2026
Merged

update mteb script#4265
pgladkows merged 1 commit into
mainfrom
update-mteb-script

Conversation

@pgladkows

Copy link
Copy Markdown
Collaborator

🛠 Summary

JIRA/Issue if applicable.
Describe the changes.

🧪 Checklist

  • Unit tests added.
  • The documentation updated.
  • Change follows security best practices.
    ``

Copilot AI review requested due to automatic review settings June 1, 2026 10:03

Copilot AI 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.

Pull request overview

This PR updates the ovms_mteb.py demo script used to run MTEB benchmarks against the OVMS embeddings endpoint, adding more CLI configurability and printing a concise score summary.

Changes:

  • Adds new CLI options for results output directory, verbosity, overwrite behavior, and evaluation splits.
  • Passes the selected eval_splits to task creation and makes evaluation.run(...) parameters configurable via CLI.
  • Prints a compact per-task score summary after running evaluation.

Comment on lines 25 to +27
parser.add_argument('--service_url', required=False, default='http://localhost:8000/v3/embeddings',
help='Specify url to embeddings endpoint. default:http://localhost:8000/v3/embeddings', dest='service_url')
parser.add_argument('--model_name', default='Alibaba-NLP/gte-large-en-v1.5', help='Model name to query. default: Alibaba-NLP/gte-large-en-v1.5',
help='Specify url to embeddings endpoint.',
dest='service_url')
parser.add_argument('--results_dir', default='results', help='Results directory. default: results',
dest='results_dir')
parser.add_argument('--verbosity', choices=['0', '1', '2', '3'], default='3', )
parser.add_argument('--overwrite_results', action='store_true', default=True, )
@pgladkows pgladkows merged commit afe5fbc into main Jun 1, 2026
2 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