Skip to content

add bge-m3 embedding dimension alias#767

Open
TurboTheTurtle wants to merge 1 commit intoCortexReach:masterfrom
TurboTheTurtle:fix-bge-m3-model-alias
Open

add bge-m3 embedding dimension alias#767
TurboTheTurtle wants to merge 1 commit intoCortexReach:masterfrom
TurboTheTurtle:fix-bge-m3-model-alias

Conversation

@TurboTheTurtle
Copy link
Copy Markdown
Contributor

Summary

Allow users to configure Huawei/BAAI BGE-M3 with the plain bge-m3 model name without also having to set embedding.dimensions manually.

  • Add bge-m3 as a 1024-dimension alias alongside the existing BAAI/bge-m3 entry.
  • Extend the embedder regression test so both common model identifiers resolve to 1024 dimensions.

Fixes #339

Validation

  • node test/embedder-error-hints.test.mjs
  • git diff --check

@rwmjhb
Copy link
Copy Markdown
Collaborator

rwmjhb commented May 9, 2026

Please rebase this branch on current master and rerun CI before merge.

The bge-m3 alias change is small and useful, but this branch is behind current master and currently reports packaging/core failures. A fresh rebase should keep this as a minimal embedder/test-only diff.

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.

Unsupported embedding model: bge-m3. Either add it to EMBEDDING_DIMENSIONS or set embedding.dimensions in config

2 participants