●Change8

Migrating to sentence-transformers v5.6.1

Version v5.6.1 introduces 1 breaking change. This guide details how to update your code.

Released: 7/23/2026

1
Breaking Changes
1
Migration Steps
3
Affected Symbols

⚠️ Check Your Code

If you use any of these symbols, you need to read this guide:

Transformer module (when flattening inputs with flash attention)position_idsposition_embeddings

Breaking Changes

●Issue #1

Users who encoded text with flash attention requested on `transformers` v5 with a RoBERTa-family checkpoint will experience silently degraded embeddings. To fix this, re-encode the corpus after upgrading to v5.6.1, as pre-fix embeddings score notably worse and should not be mixed with post-fix embeddings.

Migration Steps

  1. 1
    If you encoded text with flash attention requested on `transformers` v5 with a RoBERTa-family checkpoint, re-encode your corpus after upgrading to v5.6.1.

Release Summary

This patch release fixes silently degraded embeddings for RoBERTa-family models when flash attention is used with `transformers` v5. The issue affected specific multilingual embedding models and has been resolved by correcting `position_ids` offsetting.

Need More Details?

View the full release notes and all changes for sentence-transformers v5.6.1.

View Full Changelog