Tim Ren 0269900c60 fix: use asyncio.to_thread for embedding calls in search endpoints ... Wrap synchronous embedding_model.embed() calls with asyncio.to_thread() in both vector_search and hybrid_search methods. This prevents blocking the asyncio event loop during embedding computation, improving server responsiveness under concurrent load. Fixes #794 Signed-off-by: Tim Ren <137012659+xr843@users.noreply.github.com>		2026-03-15 16:21:19 +08:00
..
__init__.py	fix(backend): rename retriver directory to retriever	2025-12-06 22:21:30 +05:30
chunks_hybrid_search.py	fix: use asyncio.to_thread for embedding calls in search endpoints	2026-03-15 16:21:19 +08:00
documents_hybrid_search.py	feat: implement batch unread counts for notifications to reduce API calls and improve performance	2026-03-10 01:26:37 -07:00