A prompt buyers ask AI

How do search platforms actually combine keyword and semantic retrieval, and which ones let me tune the balance between the two?

Measured in 1 answer, last on Nov 18, 2026. Only the passages that name a brand are shown.

Brands in the answers

BrandNamed inRecommended inPositiveAvg. position
Algolia100%0%0%—
OpenSearch100%0%100%—
Elasticsearch100%0%100%—
Azure AI Search100%0%0%—
Weaviate100%0%100%—
Pinecone100%0%0%—
Vespa100%0%100%—

Where each answer names a brand

Answer 1 · Nov 18, 2026 · 5 sources cited
Algoliamixed
“Algolia | Primarily lexical; neural/AI ranking features | Less of a traditional BM25/vector knob”
− limited traditional BM25/vector tuning

“Algolia | Primarily lexical; neural/AI ranking features | Less of a traditional BM25/vector knob”

OpenSearchpositive
“OpenSearch | Score normalization + weighted combination, or RRF | Yes — strongly”
+ explicit control over normalization and fusion

“OpenSearch gives you unusually explicit control.”

Elasticsearchpositive
“Elasticsearch | RRF or weighted/linear fusion | Yes — strongly”
+ tunable keyword-semantic balance

“Elasticsearch | RRF or weighted/linear fusion | Yes — strongly”

+ supports RRF and weighted/convex combinations

“Elasticsearch supports RRF as well as weighted/convex combinations.”

Azure AI Search
“Azure AI Search | RRF + vector weighting | Yes, though differently”
+ vector weighting and semantic ranking controls

“Azure AI Search is slightly different: its standard hybrid retrieval uses RRF, but you can give vector queries/fields a weight, thereby changing the vector component's influence.”

Weaviatepositive
“Weaviate | Score fusion with configurable alpha | Yes — directly”
+ configurable relative weighting and fusion strategies

“Weaviate exposes the concept very directly: hybrid search combines BM25 and vector results, and the relative weighting is configurable.”

Pinecone
“Pinecone | Hybrid sparse+dense retrieval | Yes, via sparse/dense weighting”
Vespapositive
“Vespa | Custom ranking expressions | Yes — extremely flexible”
+ flexibility of custom ranking expressions

“Vespa | Custom ranking expressions | Yes — extremely flexible”

Sources the answers cite

← All prompts and the Site Search & Personalization ranking