Navigate a neighbour graph
A useful starting point for similarity search. Tune the graph and search breadth, then measure the recall your application needs.
Metadata filtering: candidate post-filtering All products Find similar meaning
Explore how a question finds nearby meaningIllustrative example
Vector search
Find similar meaning
in your data.
Store embeddings, retrieve nearby matches and connect each result to its application record.
EXPLORE A QUERY
Meaning has neighbours.Conceptual embedding space
Query vector Similar candidates
A simplified projection. Search uses the full embedding, not this picture.
YOUR QUESTION
“Something warm for a mountain hike”
Similarity search → record IDs
- 01Insulated trail jacket
- 02Lightweight fleece layer
- 03All-weather hiking shell
YOUR QUESTION
“A quiet weekend close to nature”
Similarity search → record IDs
- 01Lakeside cabin escapes
- 02Forest walks near the city
- 03A weekend in the mountains
YOUR QUESTION
Illustrative matches · no live query “How do I get back into my account?”
Similarity search → record IDs
- 01Reset your password
- 02Recover account access
- 03Update your sign-in details
Embed your data. Search by similarity.
Find relevant content even when a query uses different words. Your application supplies the embeddings; OriginChainDB indexes and searches them.
Store your embeddings
Generate vectors with your model and write them with record IDs. Include the metadata you need for filtering at write time.
Retrieve useful matches
Submit a query embedding, choose a distance metric and fetch the records behind the returned IDs. Keep model and dimensions consistent.
Metadata filters use exact equality. Filtering behaviour varies by index; IVF-PQ filtering is not supported.
Try a metadata filterCHOOSE THE TRADE-OFFYour workload.
Your workload.
Your index.
Memory, query effort and recall pull in different directions. Choose with measurements from your own dataset.
Search selected partitions
Organize vectors into cells and control how many are probed. More probes inspect more candidates, with a corresponding query cost.
Metadata filtering: during cell scansCompress the representation
Evaluate a smaller vector representation when memory matters. Test the effect of compression on your data and recall target.
Metadata filtering: not supportedDense rankingA · C · BEmbedding similarity
Sparse rankingB · A · DSparse-vector similarity
Reciprocal rank fusionA · B · C · DIllustrative ordering · equal weights
HYBRID VECTOR RETRIEVALTwo representations.
Two representations.
One ranked list.
Write dense and sparse vectors under matching IDs. The hybrid endpoint combines their rankings so your application can retrieve one set of candidates.
This endpoint fuses vector rankings. Full-text BM25 is a separate search path.
Build a hybrid queryWhy vectors on OriginChainDB?
Context beside similarity
Use shared IDs across records, vectors, graph relationships and searchable text to assemble the context your application needs.
Explore retrieval patternsIndex choices
Start with HNSW, then evaluate IVF or IVF-PQ when your workload calls for a different memory and recall trade-off.
Compare vector indexesDense and sparse together
Use the hybrid vector endpoint to combine both rankings with reciprocal rank fusion and return one list of matches.
Try hybrid retrievalPut it to work.
Choose an index for your workload
Review index setup, filtering behaviour and memory requirements. Measure recall on your own data before changing the default.
Vector index guideWork with vector records
Use the dedicated vector record API for indexes, namespaces and supplied vectors with metadata.
Vector record APIEmbed an existing table
On dedicated configurations, use a saved embedder and an explicit row limit to generate vectors from selected text.
Embedding runsCombine dense and sparse vectors
Write both representations under the same ID and fuse their ranked results through the hybrid top-k endpoint.
Hybrid retrieval example