Merge pull request #16523 from expruc/perf/hybrid_search_bm_25

perf: disable collection retrieval and bm_25 calculation if bm_25 weight is 0

backend/open_webui/retrieval/utils.py

@@ -124,6 +124,8 @@ def query_doc_with_hybrid_search(
     hybrid_bm25_weight: float,
 ) -> dict:
     try:
+        # BM_25 required only if weight is greater than 0
+        if hybrid_bm25_weight > 0:
             log.debug(f"query_doc_with_hybrid_search:doc {collection_name}")
             bm25_retriever = BM25Retriever.from_texts(
                 texts=collection_result.documents[0],
@@ -337,6 +339,8 @@ def query_collection_with_hybrid_search(
     # Fetch collection data once per collection sequentially
     # Avoid fetching the same data multiple times later
     collection_results = {}
+    # Only retrieve entire collection if bm_25 calculation is required
+    if hybrid_bm25_weight > 0:
         for collection_name in collection_names:
             try:
                 log.debug(
@@ -348,7 +352,9 @@ def query_collection_with_hybrid_search(
             except Exception as e:
                 log.exception(f"Failed to fetch collection {collection_name}: {e}")
                 collection_results[collection_name] = None
-
+    else:
+        for collection_name in collection_names:
+            collection_results[collection_name] = []
     log.info(
         f"Starting hybrid search for {len(queries)} queries in {len(collection_names)} collections..."
     )