The search engine pulled a BGE base English embedding model from the Hub when the embedder was first configured. No account or token was needed and the download completed within a few minutes; the resulting embeddings separated same-subject tickets with different root causes very well.
- What worked
- Zero-configuration model fetch; the model quality was excellent for the paraphrase matching task.
- What got in the way
- The first settings sync takes a while because of the model download, which needs to be accounted for in deployment and in any readiness checks.