landing_page/qdrant-landing/content/documentation/concepts/hybrid-queries.md

title, weight, aliases, hideInSidebar

title	weight	aliases	hideInSidebar
Hybrid Queries	57	../hybrid-queries	false

Hybrid and Multi-Stage Queries

Available as of v1.10.0

With the introduction of multiple named vectors per point, there are use-cases when the best search is obtained by combining multiple queries, or by performing the search in more than one stage.

Qdrant has a flexible and universal interface to make this possible, called Query API (API reference).

The main component for making the combinations of queries possible is the prefetch parameter, which enables making sub-requests.

Specifically, whenever a query has at least one prefetch, Qdrant will:

1. Perform the prefetch query (or queries),
2. Apply the main query over the results of its prefetch(es).

Additionally, prefetches can have prefetches themselves, so you can have nested prefetches.

Using offset parameter only affects the main query. This means that the prefetches must have a limit of at least limit + offset of the main query, otherwise you can get an empty result.

Hybrid Search

One of the most common problems when you have different representations of the same data is to combine the queried points for each representation into a single result.

{{< figure src="/docs/fusion-idea.png" caption="Fusing results from multiple queries" width="80%" >}}

For example, in text search, it is often useful to combine dense and sparse vectors to get the best of both worlds: semantic understanding from dense vectors and precise word matching from sparse vectors.

Qdrant has a few ways of fusing the results from different queries: rrf and dbsf

Reciprocal Rank Fusion (RRF)

RRF considers the positions of results within each query and boosts those that appear closer to the top in multiple sets of results. The score of a document is calculated using its rank in each result set: score(d\in D) = \sum_{r_d\in R(d)} \frac{1}{k + \frac{r_d + 1}{w_r} - 1}

Where:

• D the set of points across all results
• R(d) is the set of rankings for a particular document
• k is a constant (set to 2 by default)
• r is an ordered set of results from one source
• r_d is the rank of document d in ranking r
• w_r is the weight of ranking r (set to 1 by default)

Because w_r defaults to 1, without setting explicit weights, the formula can be simplified to the original RRF function:

score(d\in D) = \sum_{r_d\in R(d)} \frac{1}{k + r_d}

Here is an example of RRF for a query containing two prefetches against different named vectors configured to hold sparse and dense vectors, respectively.

{{< code-snippet path="/documentation/headless/snippets/query-points/hybrid-rrf/" >}}

Setting RRF Constant k

Available as of v1.16.0

To change the value of constant k in the formula, use the dedicated rrf query.

{{< code-snippet path="/documentation/headless/snippets/query-points/hybrid-rrf-k/" >}}

Weighted RRF

Available as of v1.17.0

By default, each query is assigned an equal weight. In reality, some queries are stronger, more discriminative, or more domain-specific than others. For example, a semantic search model understands meaning better than a simple keyword matcher. Assigning equal weight to both can cause the weaker model to negatively influence results, leading to a suboptimal search experience. To address this, you can assign greater weight to rankers that perform well.

The rrf query allows you to configure relative weights for each of the prefetches. For example, if you have two prefetches and assign a weight of 3.0 to the first and 1.0 to the second, a document ranked third in the first query scores the same as a document ranked first in the second query. In the case of non-overlapping result sets, these weights return three results from the first set for every one result from the second set.

Weights should be provided as an array of numbers, where each weight is applied to the corresponding prefetch in the order they are defined. The number of weights must match the number of prefetches.

{{< code-snippet path="/documentation/headless/snippets/query-points/hybrid-rrf-weights/" >}}

Distribution-Based Score Fusion (DBSF)

Available as of v1.11.0

DBSF normalizes the scores of the points in each query, using the mean +/- the 3rd standard deviation as limits, and then sums the scores of the same point across different queries.

dbsf is stateless and calculates the normalization limits only based on the results of each query, not on all the scores that it has seen.

Multi-stage queries

In general, larger vector representations give more accurate search results, but makes them more expensive to compute.

Splitting the search into two stages is a known technique to mitigate this effect:

• First, use a smaller and cheaper representation to get a large list of candidates.
• Then, re-score the candidates using the larger and more accurate representation.

There are a few ways to build search architectures around this idea:

• The quantized vectors as a first stage, and the full-precision vectors as a second stage.
• Leverage Matryoshka Representation Learning (MRL) to generate candidate vectors with a shorter vector, and then refine them with a longer one.
• Use regular dense vectors to pre-fetch the candidates, and then re-score them with a multi-vector model like ColBERT.

To get the best of all worlds, Qdrant has a convenient interface to perform the queries in stages, such that the coarse results are fetched first, and then they are refined later with larger vectors.

Re-scoring examples

Fetch 1000 results using a shorter MRL byte vector, then re-score them using the full vector and get the top 10.

{{< code-snippet path="/documentation/headless/snippets/query-points/hybrid-rescoring/" >}}

Fetch 100 results using the default vector, then re-score them using a multi-vector to get the top 10.

{{< code-snippet path="/documentation/headless/snippets/query-points/hybrid-rescoring-multivector/" >}}

It is possible to combine all the above techniques in a single query:

{{< code-snippet path="/documentation/headless/snippets/query-points/hybrid-rescoring-multistage/" >}}

Grouping

Available as of v1.11.0

It is possible to group results by a certain field. This is useful when you have multiple points for the same item, and you want to avoid redundancy of the same item in the results.

REST API (Schema):

{{< code-snippet path="/documentation/headless/snippets/query-groups/basic/" >}}

For more information on the grouping capabilities refer to the reference documentation for search with grouping and lookup.