Search operators

This page covers the search operators that can be used in queries, such as vector search operators (nearText, nearVector, nearObject, etc), keyword search operator (bm25), hybrid search operator (hybrid).

Only one search operator can be added to queries on the collection level.

Operator availability

Built-in operators

These operators are available in all Weaviate instances regardless of configuration.

nearVector
nearObject
hybrid
bm25

Module-specific operators

Module-specific search operators are made available in certain Weaviate modules.

By adding relevant modules, you can use the following operators:

nearText
Multimodal search
ask

Vector search operators

nearXXX operators allow you to find data objects based on their vector similarity to the query. They query can be a raw vector (nearVector) or an object UUID (nearObject).

If the appropriate vectorizer model is enabled, a text query (nearText), an image (nearImage), or another media input may be be used as the query.

All vector search operators can be used with a certainty or distance threshold specified, as well as a limit operator or an autocut operator to specify the desired similarity or distance between the query and the results

nearVector

nearVector finds data objects closest to an input vector.

Variables

Variable	Required	Type	Description
`vector`	yes	`[float]`	This variable takes a vector embedding in the form of an array of floats. The array should have the same length as the vectors in this collection.
`distance`	no	`float`	The maximum allowed distance to the provided search input. Cannot be used together with the `certainty` variable. The interpretation of the value of the distance field depends on the distance metric used.
`certainty`	no	`float`	Normalized Distance between the result item and the search vector. Normalized to be between 0 (perfect opposite) and 1 (identical vectors). Can't be used together with the `distance` variable.

Example

Python

collection = client.collections.use("Article")

response = collection.query.near_vector(
    near_vector=query_vector,
    distance=0.7,
    limit=5,
)

for o in response.objects:
    print(o.properties)

package main

import (
  "context"
  "fmt"

  "github.com/weaviate/weaviate-go-client/v5/weaviate"
  "github.com/weaviate/weaviate-go-client/v5/weaviate/graphql"
)

func main() {
  cfg := weaviate.Config{
    Host:   "localhost:8080",
    Scheme: "http",
  }
  client, err := weaviate.NewClient(cfg)
  if err != nil {
    panic(err)
  }

  className := "Publication"

  name := graphql.Field{Name: "name"}
  _additional := graphql.Field{
    Name: "_additional", Fields: []graphql.Field{
      {Name: "certainty"},  // only supported if distance==cosine
      {Name: "distance"},   // always supported
    },
  }

  nearVector := client.GraphQL().NearVectorArgBuilder().
    WithVector([]float32{0.1, -0.15, 0.3})  // Replace with a compatible vector

  ctx := context.Background()
  result, err := client.GraphQL().Get().
    WithClassName(className).
    WithFields(name, _additional).
    WithNearVector(nearVector).
    Do(ctx)

  if err != nil {
    panic(err)
  }
  fmt.Printf("%v", result)
}

Replace the placeholder vector with a compatible vector.

Bash

# Note: under _additional, `certainty` is only supported if distance==cosine, but `distance` is always supported

echo '{
  "query": "{
    Get {
      Publication(
        nearVector: {
          vector: [0.1, -0.15, 0.3]
        }
      ) {
        name
        _additional {
          certainty
          distance
        }
      }
    }
  }"
}' | curl \
    -X POST \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Bearer learn-weaviate' \
    -d @- \
    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{
  Get{
    Publication(
      nearVector: {
        vector: [0.1, -0.15, 0.3]  # Replace with a compatible vector
      }
    ){
      name
      _additional {
        certainty
      }
    }
  }
}

:::

nearObject

nearVector finds data objects closest to an existing object in the same collection. The object is typically specified by its UUID.

Note: You can specify an object's id or beacon in the argument, along with a desired certainty.
Note that the first result will always be the object used for search.

Variables

Variable	Required	Type	Description
`id`	yes	`UUID`	Data object identifier in the uuid format.
`beacon`	no	`url`	Data object identifier in the beacon URL format. E.g., `weaviate://<hostname>/<kind>/id`.
`distance`	no	`float`	The maximum allowed distance to the provided search input. Cannot be used together with the `certainty` variable. The interpretation of the value of the distance field depends on the distance metric used.
`certainty`	no	`float`	Normalized Distance between the result item and the search vector. Normalized to be between 0 (perfect opposite) and 1 (identical vectors). Can't be used together with the `distance` variable.

Example

Python

collection = client.collections.use("Article")

response = collection.query.near_object(
    near_object=object_id,
    distance=0.6,
    limit=5,
)

for o in response.objects:
    print(o.properties)

package main

import (
  "context"
  "fmt"

  "github.com/weaviate/weaviate-go-client/v5/weaviate"
  "github.com/weaviate/weaviate-go-client/v5/weaviate/graphql"
)

func main() {
  cfg := weaviate.Config{
    Host:   "localhost:8080",
    Scheme: "http",
  }
  client, err := weaviate.NewClient(cfg)
  if err != nil {
    panic(err)
  }

  className := "Publication"
  fields := []graphql.Field{
    {Name: "name"},
    {Name: "_additional", Fields: []graphql.Field{
      {Name: "certainty"}, // certainty only supported if distance==cosine
      {Name: "distance"},  // distance always supported
    }},
  }
  nearObject := client.GraphQL().NearObjectArgBuilder().WithID("32d5a368-ace8-3bb7-ade7-9f7ff03eddb6")
  ctx := context.Background()

  result, err := client.GraphQL().Get().
    WithClassName(className).
    WithFields(fields...).
    WithNearObject(nearObject).
    Do(ctx)

  if err != nil {
    panic(err)
  }
  fmt.Printf("%v", result)
}

Curl

# Note: prior to v1.14, use `certainty` instead of `distance`
# Under _additional, `certainty` is only supported if distance==cosine, but `distance` is always supported
echo '{
  "query": "{
    Get {
      Publication(
        nearObject: {
          id: \"32d5a368-ace8-3bb7-ade7-9f7ff03eddb6\",
          distance: 0.6
        }
      ) {
        name
        _additional {
          certainty
          distance
        }
      }
    }
  }"
}' | curl \
    -X POST \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Bearer learn-weaviate' \
    -d @- \
    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{
  Get{
    Publication(
      nearObject: {
        id: "32d5a368-ace8-3bb7-ade7-9f7ff03eddb6",  # or weaviate://localhost/32d5a368-ace8-3bb7-ade7-9f7ff03eddb6
        distance: 0.6  # prior to v1.14, use certainty: 0.7
      }
    ) {
      name
      _additional {
        certainty  # only works if distance==cosine
        distance   # always works
      }
    }
  }
}

Expected response

JSON

{
  "data": {
    "Get": {
      "Publication": [
        {
          "_additional": {
            "distance": -1.1920929e-07
          },
          "name": "The New York Times Company"
        },
        {
          "_additional": {
            "distance": 0.059879005
          },
          "name": "New York Times"
        },
        {
          "_additional": {
            "distance": 0.09176409
          },
          "name": "International New York Times"
        },
        {
          "_additional": {
            "distance": 0.13954824
          },
          "name": "New Yorker"
        },
        ...
      ]
    }
  }
}

nearText

The nearText operator finds data objects based on their vector similarity to a natural language query.

This operator is enabled if a compatible vectorizer module is configured for the collection. Compatible vectorizer modules are:

Any text2vec module
Any multi2vec module

Variables

Variable	Required	Type	Description
`concepts`	yes	`[string]`	An array of strings that can be natural language queries, or single words. If multiple strings are used, a centroid is calculated and used. Learn more about how the concepts are parsed here.
`distance`	no	`float`	The maximum allowed distance to the provided search input. Cannot be used together with the `certainty` variable. The interpretation of the value of the distance field depends on the distance metric used.
`certainty`	no	`float`	Normalized Distance between the result item and the search vector. Normalized to be between 0 (perfect opposite) and 1 (identical vectors). Can't be used together with the `distance` variable.
`autocorrect`	no	`boolean`	Autocorrect input text values. Requires the `text-spellcheck` module to be present & enabled.
`moveTo`	no	`object{}`	Move your search term closer to another vector described by keywords
`moveTo{concepts}`	no	`[string]`	An array of strings - natural language queries or single words. If multiple strings are used, a centroid is calculated and used.
`moveTo{objects}`	no	`[UUID]`	Object IDs to move the results to. This is used to "bias" NLP search results into a certain direction in vector space.
`moveTo{force}`	no	`float`	The force to apply to a particular movement. Must be between 0 and 1 where 0 is equivalent to no movement and 1 is equivalent to largest movement possible.
`moveAwayFrom`	no	`object{}`	Move your search term away from another vector described by keywords
`moveAwayFrom{concepts}`	no	`[string]`	An array of strings - natural language queries or single words. If multiple strings are used, a centroid is calculated and used.
`moveAwayFrom{objects}`	no	`[UUID]`	Object IDs to move the results from. This is used to "bias" NLP search results into a certain direction in vector space.
`moveAwayFrom{force}`	no	`float`	The force to apply to a particular movement. Must be between 0 and 1 where 0 is equivalent to no movement and 1 is equivalent to largest movement possible.

Example I

This example shows an example usage the nearText operator, including how to bias results towards another search query.

Python

# START-ANY
import weaviate
import weaviate.classes as wvc
from weaviate.classes.query import Move
import os

client = weaviate.connect_to_local()

# END-ANY

# Actual client instantiation
client.close()

from weaviate.classes.init import Auth

# Best practice: store your credentials in environment variables
weaviate_url = os.environ["WEAVIATE_URL"]
weaviate_api_key = os.environ["WEAVIATE_API_KEY"]
openai_api_key = os.environ["OPENAI_API_KEY"]

client = weaviate.connect_to_weaviate_cloud(
    cluster_url=weaviate_url,
    auth_credentials=Auth.api_key(weaviate_api_key),
    headers={
        "X-OpenAI-Api-Key": openai_api_key,
    }
)

# START-ANY
try:
# END-ANY

# START-ANY
    publications = client.collections.use("Publication")

    response = publications.query.near_text(
        query="fashion",
        distance=0.6,
        move_to=Move(force=0.85, concepts="haute couture"),
        move_away=Move(force=0.45, concepts="finance"),
        return_metadata=wvc.query.MetadataQuery(distance=True),
        limit=2
    )

    for o in response.objects:
        print(o.properties)
        print(o.metadata)
# END-ANY

# START-ANY

finally:
    client.close()
# END-ANY

package main

import (
  "context"
  "fmt"

  "github.com/weaviate/weaviate-go-client/v5/weaviate"
  "github.com/weaviate/weaviate-go-client/v5/weaviate/graphql"
)

func main() {
  cfg := weaviate.Config{
    Host:   "localhost:8080",
    Scheme: "http",
  }
  client, err := weaviate.NewClient(cfg)
  if err != nil {
    panic(err)
  }

  className := "Publication"

  name := graphql.Field{Name: "name"}
  _additional := graphql.Field{
    Name: "_additional", Fields: []graphql.Field{
      {Name: "certainty"}, // only supported if distance==cosine
      {Name: "distance"},  // always supported
    },
  }

  concepts := []string{"fashion"}
  distance := float32(0.6)
  moveAwayFrom := &graphql.MoveParameters{
    Concepts: []string{"finance"},
    Force:    0.45,
  }
  moveTo := &graphql.MoveParameters{
    Concepts: []string{"haute couture"},
    Force:    0.85,
  }
  nearText := client.GraphQL().NearTextArgBuilder().
    WithConcepts(concepts).
    WithDistance(distance). // use WithCertainty(certainty) prior to v1.14
    WithMoveTo(moveTo).
    WithMoveAwayFrom(moveAwayFrom)

  ctx := context.Background()

  result, err := client.GraphQL().Get().
    WithClassName(className).
    WithFields(name, _additional).
    WithNearText(nearText).
    Do(ctx)

  if err != nil {
    panic(err)
  }
  fmt.Printf("%v", result)
}

Curl

# Note: Under nearText, use `certainty` instead of distance prior to v1.14
# Under _additional, `certainty` is only supported if distance==cosine, but `distance` is always supported
echo '{
  "query": "{
    Get {
      Publication(
        nearText: {
          concepts: [\"fashion\"],
          distance: 0.6,
          moveAwayFrom: {
            concepts: [\"finance\"],
            force: 0.45
          },
          moveTo: {
            concepts: [\"haute couture\"],
            force: 0.85
          }
        }
      ) {
        name
        _additional {
          certainty
          distance
        }
      }
    }
  }"
}' | curl \
    -X POST \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Bearer learn-weaviate' \
    -H "X-OpenAI-Api-Key: $OPENAI_API_KEY" \
    -d @- \
    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{
  Get{
    Publication(
      nearText: {
        concepts: ["fashion"],
        distance: 0.6 # prior to v1.14 use "certainty" instead of "distance"
        moveAwayFrom: {
          concepts: ["finance"],
          force: 0.45
        },
        moveTo: {
          concepts: ["haute couture"],
          force: 0.85
        }
      }
    ){
      name
      _additional {
        certainty # only supported if distance==cosine.
        distance  # always supported
      }
    }
  }
}

Example II

You can also bias results toward other data objects. For example, in this query, we move our query about "travelling in asia", towards an article on food.

Python

collection = client.collections.use("Article")

response = collection.query.near_text(
    query="travelling in Asia",
    certainty=0.7,
    move_to=wvc.query.Move(
        force=0.75,
        objects="c4209549-7981-3699-9648-61a78c2124b9"
    ),
    return_metadata=wvc.query.MetadataQuery(certainty=True),
    limit=5,
)

for o in response.objects:
    print(o.properties)
    print(o.metadata.certainty)

package main

import (
  "context"
  "fmt"

  "github.com/weaviate/weaviate-go-client/v5/weaviate"
  "github.com/weaviate/weaviate-go-client/v5/weaviate/graphql"
)

func main() {
  cfg := weaviate.Config{
    Host:   "localhost:8080",
    Scheme: "http",
  }
  client, err := weaviate.NewClient(cfg)
  if err != nil {
    panic(err)
  }

  className := "Article"

  title := graphql.Field{Name: "title"}
  summary := graphql.Field{Name: "summary"}
  _additional := graphql.Field{
    Name: "_additional", Fields: []graphql.Field{
      {Name: "certainty"},
    },
  }

  concepts := []string{"travelling in Asia"}
  certainty := float32(0.7)
  moveTo := &graphql.MoveParameters{
    Objects: []graphql.MoverObject{
	  // this ID is of the article: "Tohoku: A Japan destination for all seasons."
      {ID: "c4209549-7981-3699-9648-61a78c2124b9"},
    },
    Force: 0.85,
  }
  nearText := client.GraphQL().NearTextArgBuilder().
    WithConcepts(concepts).
    WithCertainty(certainty).
    WithMoveTo(moveTo)

  ctx := context.Background()

  result, err := client.GraphQL().Get().
    WithClassName(className).
    WithFields(title, summary, _additional).
    WithNearText(nearText).
    Do(ctx)

  if err != nil {
    panic(err)
  }
  fmt.Printf("%v", result)
}

Curl

# The ID belongs to the article "Tohoku: A Japan destination for all seasons."

echo '{
  "query": "{
    Get {
      Article(
        nearText: {
          concepts: [\"travelling in Asia\"],
          certainty: 0.7,
          moveTo: {
            objects: [{
              id: \"c4209549-7981-3699-9648-61a78c2124b9\"
            }]
            force: 0.85
          }
        }
      ) {
        title
        summary
        _additional {
          certainty
        }
      }
    }
  }"
}' | curl \
    -X POST \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Bearer learn-weaviate' \
    -H "X-OpenAI-Api-Key: $OPENAI_API_KEY" \
    -d @- \
    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{
  Get {
    Article(
      nearText: {
        concepts: ["travelling in Asia"],
        certainty: 0.7,
        moveTo: {
          objects: [{
            # this ID is of the article:
            # "Tohoku: A Japan destination for all seasons."
            id: "c4209549-7981-3699-9648-61a78c2124b9"
          }]
          force: 0.85
        }
      }
    ) {
      title
      summary
      _additional {
        certainty
      }
    }
  }
}

Expected response

JSON

{
  "data": {
    "Get": {
      "Article": [
        {
          "_additional": {
            "certainty": 0.9619976580142975
          },
          "summary": "We've scoured the planet for what we think are 50 of the most delicious foods ever created. A Hong Kong best food, best enjoyed before cholesterol checks. When you have a best food as naturally delicious as these little fellas, keep it simple. Courtesy Matt@PEK/Creative Commons/FlickrThis best food Thai masterpiece teems with shrimp, mushrooms, tomatoes, lemongrass, galangal and kaffir lime leaves. It's a result of being born in a land where the world's most delicious food is sold on nearly every street corner.",
          "title": "World food: 50 best dishes"
        },
        {
          "_additional": {
            "certainty": 0.9297388792037964
          },
          "summary": "The look reflects the elegant ambiance created by interior designer Joyce Wang in Hong Kong, while their mixology program also reflects the original venue. MONO Hong Kong , 5/F, 18 On Lan Street, Central, Hong KongKoral, The Apurva Kempinski Bali, IndonesiaKoral's signature dish: Tomatoes Bedugul. Esterre at Palace Hotel TokyoLegendary French chef Alain Ducasse has a global portfolio of restaurants, many holding Michelin stars. John Anthony/JW Marriott HanoiCantonese cuisine from Hong Kong is again on the menu, this time at the JW Marriott in Hanoi. Stanley takes its name from the elegant Hong Kong waterside district and the design touches reflect this legacy with Chinese antiques.",
          "title": "20 best new Asia-Pacific restaurants to try in 2020"
        }
        ...
      ]
    }
  }
}

Additional information

Concept parsing

A nearText query will interpret each term in an array input as distinct strings to be vectorized. If multiple strings are passed, the query vector will be an average vector of the individual string vectors.

["New York Times"] = one vector position is determined based on the occurrences of the words
["New", "York", "Times"] = all concepts have a similar weight.
["New York", "Times"] = a combination of the two above.

A practical example would be: concepts: ["beatles", "John Lennon"]

Semantic Path

Only available in txt2vec-contextionary module

The semantic path returns an array of concepts from the query to the data object. This allows you to see which steps Weaviate took and how the query and data object are interpreted.

Property	Description
`concept`	the concept that is found in this step.
`distanceToNext`	the distance to the next step (null for the last step).
`distanceToPrevious`	this distance to the previous step (null for the first step).
`distanceToQuery`	the distance of this step to the query.
`distanceToResult`	the distance of the step to this result.

Note: Building a semantic path is only possible if a nearText: {} operator is set as the explore term represents the beginning of the path and each search result represents the end of the path. Since nearText: {} queries are currently exclusively possible in GraphQL, the semanticPath is therefore not available in the REST API.

Example: showing a semantic path without edges.

Python

# Semantic path is not yet supported by the V4 client. Please use a raw GraphQL query instead.
response = client.graphql_raw_query(
"""
{
  Get {
    Publication (
      nearText:{
        concepts: ["fashion"],
        distance: 0.6,
        moveAwayFrom: {
          concepts: ["finance"],
          force: 0.45
        },
        moveTo: {
          concepts: ["haute couture"],
          force: 0.85
        }
      }
    ) {
      name
      _additional {
        semanticPath {
          path {
            concept
            distanceToNext
            distanceToPrevious
            distanceToQuery
            distanceToResult
          }
        }
      }
    }
  }
}
"""
)

package main

import (
  "context"
  "fmt"

  "github.com/weaviate/weaviate-go-client/v5/weaviate"
  "github.com/weaviate/weaviate-go-client/v5/weaviate/graphql"
)

func main() {
  cfg := weaviate.Config{
    Host:   "localhost:8080",
    Scheme: "http",
  }
  client, err := weaviate.NewClient(cfg)
  if err != nil {
    panic(err)
  }

  className := "Publication"
  fields := []graphql.Field{
    {Name: "name"},
    {Name: "_additional", Fields: []graphql.Field{
      {Name: "semanticPath", Fields: []graphql.Field{
        {Name: "path", Fields: []graphql.Field{
          {Name: "concept"},
          {Name: "distanceToNext"},
          {Name: "distanceToPrevious"},
          {Name: "distanceToQuery"},
          {Name: "distanceToResult"},
        }},
      }},
    }},
  }

  concepts := []string{"fashion"}
  moveTo := &graphql.MoveParameters{
    Concepts: []string{"haute couture"},
    Force:    0.85,
  }
  moveAwayFrom := &graphql.MoveParameters{
    Concepts: []string{"finance"},
    Force:    0.45,
  }
  nearText := client.GraphQL().NearTextArgBuilder().
    WithConcepts(concepts).
    WithDistance(0.6). // prior to v1.14, use WithCertainty(0.7)
    WithMoveTo(moveTo).
    WithMoveAwayFrom(moveAwayFrom)

  ctx := context.Background()

  result, err := client.GraphQL().Get().
    WithClassName(className).
    WithFields(fields...).
    WithNearText(nearText).
    Do(ctx)

  if err != nil {
    panic(err)
  }
  fmt.Printf("%v", result)
}

Curl

# Note: Under nearText, use `certainty` instead of `distance` prior to v1.14

echo '{
  "query": "{
    Get {
      Publication (
        nearText: {
          concepts: [\"fashion\"],
          distance: 0.6,
          moveAwayFrom: {
            concepts: [\"finance\"],
            force: 0.45
          },
          moveTo: {
            concepts: [\"haute couture\"],
            force: 0.85
          }
        }
      ) {
        name
        _additional {
          semanticPath{
            path {
              concept
              distanceToNext
              distanceToPrevious
              distanceToQuery
              distanceToResult
            }
          }
        }
      }
    }
  }"
}' | curl \
    -X POST \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Bearer learn-weaviate' \
    -H "X-OpenAI-Api-Key: $OPENAI_API_KEY" \
    -d @- \
    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{
  Get {
    Publication (
      nearText:{
        concepts: ["fashion"],
        distance: 0.6,  # prior to v1.14 use certainty: 0.7
        moveAwayFrom: {
          concepts: ["finance"],
          force: 0.45
        },
        moveTo: {
          concepts: ["haute couture"],
          force: 0.85
        }
      }
    ) {
      name
      _additional {
        semanticPath {
          path {
            concept
            distanceToNext
            distanceToPrevious
            distanceToQuery
            distanceToResult
          }
        }
      }
    }
  }
}

Multimodal search

Depending on the vectorizer module, you can use additional modalities such as images, audio, or video as the query, and retrieve corresponding, compatible objects.

Some modules, such as multi2vec-clip and multi2vec-bind allow you to search across modalities. For example, you can search for images using a text query, or search for text using an image query.

For more information, see specific module pages such as these:

hybrid

This operator allows you to combine BM25 and vector search to get a "best of both worlds" type search results set.

Variables

Variables	Required	Type	Description
`query`	yes	`string`	search query
`alpha`	no	`float`	weighting for each search algorithm, default 0.75
`vector`	no	`[float]`	optional to supply your own vector
`properties`	no	`[string]`	list of properties to limit the BM25 search to, default all text properties
`fusionType`	no	`string`	the type of hybrid fusion algorithm (available from `v1.20.0`)
`bm25SearchOperator`	no	`object`	set how many of the (bm25) query tokens must be present within a single searched property for an object to be considered a match. (available from `v1.31.0`)

Notes:
- alpha can be any number from 0 to 1, defaulting to 0.75. See Alpha parameter.
  - alpha = 0 forces using a pure keyword search method (BM25)
  - alpha = 1 forces using a pure vector search method
  - alpha = 0.5 weighs the BM25 and vector methods evenly
- fusionType can be rankedFusion or relativeScoreFusion
  - relativeScoreFusion (default from v1.24) adds normalized scores of the BM25 and vector search methods
  - rankedFusion (default for v1.23 and lower) adds inverted ranks of the BM25 and vector search methods

Fusion algorithms

Ranked fusion

The rankedFusion algorithm is Weaviate's original hybrid fusion algorithm.

In this algorithm, each object is scored according to its position in the results for that search (vector or keyword). The top-ranked objects in each search get the highest scores. Scores decrease going from top to least ranked. The total score is calculated by adding the rank-based scores from the vector and keyword searches.

Relative score fusion

In relativeScoreFusion the vector search and keyword search scores are scaled between 0 and 1. The highest raw score becomes 1 in the scaled scores. The lowest value is assigned 0. The remaining values are ranked between 0 and 1. The total score is a scaled sum of the normalized vector similarity and normalized BM25 scores.

Fusion scoring comparison

This example uses a small search result set to compare the ranked fusion and relative fusion algorithms. The table shows the following information:

document id, from 0 to 4
keyword score, sorted
vector search score, sorted

Search Type	(id): score	(id): score	(id): score	(id): score	(id): score
Keyword	(1): 5	(0): 2.6	(2): 2.3	(4): 0.2	(3): 0.09
Vector	(2): 0.6	(4): 0.598	(0): 0.596	(1): 0.594	(3): 0.009

The ranking algorithms use these scores to derive the hybrid ranking.

Ranked fusion

The score depends on the rank of the result. The score is equal to 1/(RANK + 60):

Search Type	(id): score	(id): score	(id): score	(id): score	(id): score
Keyword	(1): 0.0154	(0): 0.0160	(2): 0.0161	(4): 0.0167	(3): 0.0166
Vector	(2): 0.016502	(4): 0.016502	(0): 0.016503	(1): 0.016503	(3): 0.016666

As you can see, the results of each rank is identical, regardless of the input score.

Relative score fusion

Here, we normalize the scores – the largest score is set to 1 and the lowest to 0, and all entries in-between are scaled according to their relative distance to the maximum and minimum values.

Search Type	(id): score	(id): score	(id): score	(id): score	(id): score
Keyword	(1): 1.0	(0): 0.511	(2): 0.450	(4): 0.022	(3): 0.0
Vector	(2): 1.0	(4): 0.996	(0): 0.993	(1): 0.986	(3): 0.0

Here, the scores reflect the relative distribution of the original scores. For example, the vector search scores of the first 4 documents were almost identical, which is still the case for the normalized scores.

Weighting & final scores

Before adding these scores up, they are weighted according to the alpha parameter. Let’s assume alpha=0.5, meaning both search types contribute equally to the final result and therefore each score is multiplied by 0.5.

Now, we can add the scores for each document up and compare the results from both fusion algorithms.

Algorithm Type	(id): score	(id): score	(id): score	(id): score	(id): score
Ranked	(2): 0.016301	(1): 0.015952	(0): 0.015952	(4): 0.016600	(3): 0.016630
Relative	(1): 0.993	(0): 0.752	(2): 0.725	(4): 0.509	(3): 0.0

What can we learn from this?

For the vector search, the scores for the top 4 objects (IDs 2, 4, 0, 1) were almost identical, and all of them were good results. While for the keyword search, one object (ID 1) was much better than the rest.

This is captured in the final result of relativeScoreFusion, which identified the object ID 1 the top result. This is justified because this document was the best result in the keyword search with a big gap to the next-best score and in the top group of vector search.

In contrast, for rankedFusion, the object ID 2 is the top result, closely followed by objects ID 1 and ID 0.

For a fuller discussion of fusion methods, see this blog post

Additional metadata response

Hybrid search results are sorted by a score, derived as a fused combination of their BM25F score and nearText similarity (higher is more relevant). This score, and additionally the explainScore metadata can be optionally retrieved in the response.

Example

Python

collection = client.collections.use("Article")

response = collection.query.hybrid(
    query="Fisherman that catches salmon",
    alpha=0.5,
    return_metadata=wvc.query.MetadataQuery(score=True, explain_score=True),
    limit=5,
)

for o in response.objects:
    print(o.properties)
    print(o.metadata.score)
    print(o.metadata.explain_score)

hybrid := &HybridArgumentBuilder{}
	hybrid.WithQuery("Fisherman that catches salmon").WithAlpha(0.5)

	query := builder.WithClassName("Article").WithHybrid(hybrid).build()

Curl

echo '{
  "query": "{
    Get {
      Article(
        hybrid: {
          query: \"Fisherman that catches salmon\"
          alpha: 0.5
        }
      ) {
        title
        summary
        _additional { score explainScore }
      }
    }
  }"
}' | curl \
    -X POST \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Bearer learn-weaviate' \
    -H "X-OpenAI-Api-Key: $OPENAI_API_KEY" \
    -d @- \
    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{
  Get {
    Article (
      hybrid: {
        query: "Fisherman that catches salmon"
        alpha: 0.5
      }
    ) {
      title
      summary
      _additional { score, explainScore }
    }
  }
}

Example with vector specified

You can optionally supply the vector query to the vector variable. This will override the query variable for the vector search component of the hybrid search.

Python

collection = client.collections.use("Article")

response = collection.query.hybrid(
    query="Fisherman that catches salmon",
    vector=query_vector,
    alpha=0.5,
    return_metadata=wvc.query.MetadataQuery(score=True, explain_score=True),
    limit=5,
)

for o in response.objects:
    print(o.properties)
    print(o.metadata.score)
    print(o.metadata.explain_score)

hybrid := &HybridArgumentBuilder{}
	hybrid.WithQuery("Fisherman that catches salmon").WithVector(1, 2, 3).WithAlpha(0.5)

	query := builder.WithClassName("Article").WithHybrid(hybrid).build()

Curl

# The `vector` below is optional. Not needed if Weaviate handles the vectorization.
# If you provide your own embeddings, put the vector query there, and make sure it has the correct number of dimensions.
echo '{
  "query": "{
      Get {
        Article(
          hybrid: {
            query: \"Fisherman that catches salmon\"
            alpha: 0.5
            vector: [1, 2, 3]
          })
        {
          title
          summary
          _additional { score }
      }
    }
  }"
}' | curl \
    -X POST \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Bearer learn-weaviate' \
    -H "X-OpenAI-Api-Key: $OPENAI_API_KEY" \
    -d @- \
    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{
  Get {
    Article (
      hybrid: {
        query: "Fisherman that catches salmon"
        alpha: 0.5
        vector: [1, 2, 3] # optional. Not needed if Weaviate handles the vectorization. If you provide your own embeddings, put the vector query here.
      })
     {
      title
      summary
      _additional { score }
    }
  }
}

Hybrid with a conditional filter

A conditional (where) filter can be used with hybrid.

Python

collection = client.collections.use("Article")

response = collection.query.hybrid(
    query="How to catch an Alaskan Pollock",
    alpha=0.5,
    filters=wvc.query.Filter.by_property("wordCount").less_than(1000),
    limit=5,
)

for o in response.objects:
    print(o.properties)

where := filters.Where().WithPath([]string{"content"}).  WithOperator(filters.Equal).  WithValueString("Alaskan")  // All results must have "Alaskan" in the content propertyname = graphql.Field{Name: "summary"}hybrid := &graphql.HybridArgumentBuilder{}hybrid.WithQuery("How to catch an Alaskan Pollock").WithAlpha(0.5)resultSet, gqlErr := client.GraphQL().Get().WithClassName("Article").WithHybrid(hybrid).WithWhere(where).WithFields(name).Do(context.Background())articles := get["Article"].([]interface{})

Curl

echo '{  "query": "{      Get {        Article (          hybrid: { query: \"How to catch an Alaskan Pollock\", alpha: 0.5 }          where: { path: [\"wordCount\"], operator: LessThan, valueInt: 1000 }        ) {          title          summary      }    }  }"}' | curl \    -X POST \    -H 'Content-Type: application/json' \    -H 'Authorization: Bearer learn-weaviate' \    -H "X-OpenAI-Api-Key: $OPENAI_API_KEY" \    -d @- \    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{  Get {    Article (      hybrid: { query: "how to fish", alpha: 0.5 }      where: { path: ["wordCount"], operator: LessThan, valueInt: 1000 }    ) {      title      summary    }  }}

Specify object properties for BM25 search

A hybrid operator can accept an array of strings to limit the set of properties for the BM25 component of the search. If unspecified, all text properties will be searched.

Python

collection = client.collections.use("JeopardyQuestion")

response = collection.query.hybrid(
    query="Venus",
    alpha=0.25,
    query_properties=["question"],
    return_metadata=wvc.query.MetadataQuery(score=True),
    limit=5,
)

for o in response.objects:
    print(o.properties)
    print(o.metadata.score)

Curl

echo '{  "query": "{      Get {        JeopardyQuestion (          hybrid: {            query: \"Venus\"            alpha: 0.25            properties: [\"question\"]          }          limit: 3        )        {          question          answer          _additional { score }      }    }  }"}' | curl \    -X POST \    -H 'Content-Type: application/json' \    -H 'Authorization: Bearer learn-weaviate' \    -H "X-OpenAI-Api-Key: $OPENAI_API_KEY" \    -d @- \    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{  Get {    JeopardyQuestion(      hybrid: {        query: "Venus"        alpha: 0.25  # closer to pure keyword search        properties: ["question"]  # changing to "answer" will yield a different result set      }      limit: 3    ) {      question      answer      _additional {        score      }    }  }}

Oversearch with `relativeScoreFusion`

When relativeScoreFusion is used as the fusionType with a small search limit, a result set can be very sensitive to the limit parameter due to the normalization of the scores.

To mitigate this effect, Weaviate automatically performs a search with a higher limit (100) and then trims the results down to the requested limit.

BM25 search operator

Use bm25SearchOperator to set how the query tokens must match for an object to be considered a match in the keyword (bm25) search portion of the hybrid search. This is useful when you want to ensure that only objects with a certain number of relevant keywords are returned.

The available options are And, Or, and AndCross:

Option	Description
`And`	All of the query tokens must appear together within a single searched property. Tokens spread across different properties do not match.
`Or`	An additional parameter `minimumOrTokensMatch` must be specified, which defines how many of the query tokens must be present within a single searched property for the object to be considered a match.
`AndCross`	Every query token must be matched by at least one of the searched properties, so the tokens can be spread across different properties. All searched properties must share the same tokenization and analyzer settings, otherwise the query fails with an error. (available from `v1.38.8`)

If not set, the keyword search behaves as if Or was set with a minimumOrTokensMatch of 1.

BM25

The bm25 operator performs a keyword (sparse vector) search, and uses the BM25F ranking function to score the results. BM25F (Best Match 25 with Extension to Multiple Weighted Fields) is an extended version of BM25 that applies the scoring algorithm to multiple fields (properties), producing better results.

The search is case-insensitive, and case matching does not confer a score advantage. Stop words are removed. Stemming is not supported yet.

Schema configuration

The free parameters k1 and b are configurable and optional. See the schema reference for more details.

Variables

The bm25 operator supports the following variables:

Variables	Required	Description
`query`	yes	The keyword search query.
`properties`	no	Array of properties (fields) to search in, defaulting to all properties in the collection.
`searchOperator`	no	Set how the query tokens must match for an object to be considered a match. See Search operator for the available options. (available from `v1.31.0`)

Additional metadata response

The BM25F score metadata can be optionally retrieved in the response. A higher score indicates higher relevance.

Example query

Python

collection = client.collections.use("Article")

response = collection.query.bm25(
    query="fox",
    query_properties=["title"],
    return_metadata=wvc.query.MetadataQuery(score=True),
    limit=5,
)

for o in response.objects:
    print(o.properties)
    print(o.metadata.score)

package main

import (
  "context"
  "fmt"

  "github.com/weaviate/weaviate-go-client/v5/weaviate"
  "github.com/weaviate/weaviate-go-client/v5/weaviate/graphql"
)

func main() {
  cfg := weaviate.Config{
    Host:   "localhost:8080",
    Scheme: "http",
  }
  client, err := weaviate.NewClient(cfg)
  if err != nil {
    panic(err)
  }

  className := "Article"

  title := graphql.Field{Name: "title"}
  _additional := graphql.Field{
    Name: "_additional", Fields: []graphql.Field{
      {Name: "score"},
    },
  }

  query := "fox"
  properties := []string{"title"}
  bm25 := client.GraphQL().Bm25ArgBuilder().
    WithQuery(query).
    WithProperties(properties...)

  ctx := context.Background()

  result, err := client.GraphQL().Get().
    WithClassName(className).
    WithFields(title, _additional).
    WithBM25(bm25).
    Do(ctx)

  if err != nil {
    panic(err)
  }
  fmt.Printf("%v", result)
}

Curl

echo '{
  "query": "{
    Get {
      Article(
        bm25: {
          query: \"fox\",
          properties: [\"title\"],
        }
      ) {
        title
        _additional {
          score
        }
      }
    }
  }"
}' | curl \
    -X POST \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Bearer learn-weaviate' \
    -d @- \
    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{
  Get {
    Article(
      bm25: {
        query: "fox",
        properties: ["title"]
      }
    ) {
      title
      _additional {
        score
      }
    }
  }
}

Expected response

JSON

{
  "data": {
    "Get": {
      "Article": [
        {
          "_additional": {
            "certainty": null,
            "distance": null,
            "score": "3.4985464"
          },
          "title": "Tim Dowling: is the dog’s friendship with the fox sweet – or a bad omen?"
        }
      ]
    }
  },
  "errors": null
}

BM25 with a conditional filter

A conditional (where) filter can be used with bm25.

Python

collection = client.collections.use("Article")

response = collection.query.bm25(
    query="how to fish",
    return_metadata=wvc.query.MetadataQuery(score=True),
    filters=wvc.query.Filter.by_property("wordCount").less_than(1000),
    limit=5,
)

for o in response.objects:
    print(o.properties)
    print(o.metadata.score)

resultSet, gqlErr := client.GraphQL().Get().WithClassName("Article").WithHybrid(hybrid).WithWhere(where).WithFields(name).Do(context.Background())where := filters.Where().WithPath([]string{"wordCount"}).				WithOperator(filters.LessThan).				WithValueInt(1000)name = graphql.Field{Name: "summary"} // the output fieldbm25B := &BM25ArgumentBuilder{}bm25B = bm25B.WithQuery("How to fish").WithProperties("title", "summary")resultSet, gqlErr := client.GraphQL().Get().WithClassName("Article").WithBM25(bm25B).WithWhere(where).WithFields(name).Do(context.Background())articles := get["Article"].([]interface{})

Curl

echo '{  "query": "{    Get {      Article(        bm25: { query: \"how to fish\", properties: [\"title\"] }        where: { path: [\"wordCount\"], operator: LessThan, valueInt: 1000 }      ) {        summary        title      }    }  }"}' | curl \    -X POST \    -H 'Content-Type: application/json' \    -H 'Authorization: Bearer learn-weaviate' \    -d @- \    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{  Get {    Article (      bm25: { query: "how to fish", properties: ["title"] }      where: { path: ["wordCount"], operator: LessThan, valueInt: 1000 }    ) {      summary      title    }  }}

Expected response

JSON

{
  "data": {
    "Get": {
      "Article": [
        {
          "summary": "Sometimes, the hardest part of setting a fishing record is just getting the fish weighed. A Kentucky fisherman has officially set a new record in the state after reeling in a 9.05-pound saugeye. While getting the fish in the boat was difficult, the angler had just as much trouble finding an officially certified scale to weigh it on. In order to qualify for a state record, fish must be weighed on an officially certified scale. The previous record for a saugeye in Kentucky ws an 8 pound, 8-ounce fish caught in 2019.",
          "title": "Kentucky fisherman catches record-breaking fish, searches for certified scale"
        },
        {
          "summary": "Unpaid last month because there wasn\u2019t enough money. Ms. Hunt picks up shifts at JJ Fish & Chicken, bartends and babysits. three daughters is subsidized,and cereal fromErica Hunt\u2019s monthly budget on $12 an hourErica Hunt\u2019s monthly budget on $12 an hourExpensesIncome and benefitsRent, $775Take-home pay, $1,400Varies based on hours worked. Daycare, $600Daycare for Ms. Hunt\u2019s three daughters is subsidized, as are her electricity and internet costs. Household goods, $300Child support, $350Ms. Hunt picks up shifts at JJ Fish & Chicken, bartends and babysits to make more money.",
          "title": "Opinion | What to Tell the Critics of a $15 Minimum Wage"
        },
        ...
      ]
    }
  }
}

Search operator

Use searchOperator to set how the query tokens must match for an object to be considered a match. This is useful when you want to ensure that only objects with a certain number of relevant keywords are returned.

The available options are And, Or, and AndCross:

Option	Description
`And`	All of the query tokens must appear together within a single searched property. Tokens spread across different properties do not match.
`Or`	An additional parameter `minimumOrTokensMatch` must be specified, which defines how many of the query tokens must be present within a single searched property for the object to be considered a match.
`AndCross`	Every query token must be matched by at least one of the searched properties, so the tokens can be spread across different properties. All searched properties must share the same tokenization and analyzer settings, otherwise the query fails with an error. (available from `v1.38.8`)

If not set, the keyword search behaves as if Or was set with a minimumOrTokensMatch of 1.

An AndCross query example, restricted to question and answer because the JeopardyQuestion collection mixes tokenizations across its properties:

GraphQL

{
  Get {
    JeopardyQuestion(
      limit: 3
      bm25: {
        query: "African desert wind"
        properties: ["question", "answer"]
        searchOperator: {
          operator: AndCross
        }
      }
    ) {
      question
      answer
    }
  }
}

ask

Enabled by the module: Question Answering.

This operator allows you to return answers to questions by running the results through a Q&A model.

Variables

Variable	Required	Type	Description
`question`	yes	`string`	The question to be answered.
`certainty`	no	`float`	Desired minimal certainty or confidence of answer to the question. The higher the value, the stricter the search becomes. The lower the value, the fuzzier the search becomes. If no certainty is set, any answer that could be extracted will be returned.
`properties`	no	`[string]`	The properties of the queries collection which contains text. If no properties are set, all are considered.
`rerank`	no	`boolean`	If enabled, the qna module will rerank the result based on the answer score. For example, if the 3rd result - as determined by the previous (semantic) search contained the most likely answer, result 3 will be pushed to position 1, etc. Supported since v1.10.0

Example

Python

# QnA module use is not yet supported by the V4 client. Please use a raw GraphQL query instead.
response = client.graphql_raw_query(
"""
{
  Get {
    Article(
      ask: {
        question: "Who is the king of the Netherlands?",
        properties: ["summary"],
      },
      limit: 1
    ) {
      title
      _additional {
        answer {
          hasAnswer
          property
          result
          startPosition
          endPosition
        }
      }
    }
  }
}
"""

package main

import (
  "context"
  "fmt"

  "github.com/weaviate/weaviate-go-client/v5/weaviate"
  "github.com/weaviate/weaviate-go-client/v5/weaviate/graphql"
)

func main() {
  cfg := weaviate.Config{
    Host:   "localhost:8080",
    Scheme: "http",
  }
  client, err := weaviate.NewClient(cfg)
  if err != nil {
    panic(err)
  }

  className := "Article"
  fields := []graphql.Field{
    {Name: "title"},
    {Name: "_additional", Fields: []graphql.Field{
      {Name: "answer", Fields: []graphql.Field{
        {Name: "hasAnswer"},
        {Name: "certainty"},
        {Name: "property"},
        {Name: "result"},
        {Name: "startPosition"},
        {Name: "endPosition"},
      }},
    }},
  }
  ask := client.GraphQL().AskArgBuilder().
    WithQuestion("Who is the king of the Netherlands?").
    WithProperties([]string{"summary"})
  ctx := context.Background()

  result, err := client.GraphQL().Get().
    WithClassName(className).
    WithFields(fields...).
    WithAsk(ask).
    WithLimit(1).
    Do(ctx)

  if err != nil {
    panic(err)
  }
  fmt.Printf("%v", result)
}

Curl

echo '{
  "query": "{
    Get {
      Article(
        ask: {
          question: \"Who is the king of the Netherlands?\",
          properties: [\"summary\"]
        },
        limit: 1
      ) {
        title
        _additional {
          answer {
            hasAnswer
            property
            result
            startPosition
            endPosition
          }
        }
      }
    }
  }
  "
}' | curl \
    -X POST \
    -H 'Content-Type: application/json' \
    -H 'Authorization: Bearer learn-weaviate' \
    -H "X-OpenAI-Api-Key: $OPENAI_API_KEY" \
    -d @- \
    https://edu-demo.weaviate.network/v1/graphql

GraphQL

{
  Get {
    Article(
      ask: {
        question: "Who is the king of the Netherlands?",
        properties: ["summary"],
      },
      limit: 1
    ) {
      title
      _additional {
        answer {
          hasAnswer
          property
          result
          startPosition
          endPosition
        }
      }
    }
  }
}

Additional metadata response

The answer and a certainty can be retrieved.

Questions and feedback

Have a question or feedback? Here's how to reach us.

Community Forum

Ask questions and connect with other developers on our Community forum.

Support

Weaviate Cloud user or customer? Find the right channel on the Support page.

:::

Operator availability

Built-in operators

Module-specific operators

Vector search operators

nearVector

Variables

Example

nearObject

Variables

Example

nearText

Variables

Example I

Example II

Additional information

Concept parsing

Semantic Path

Multimodal search

hybrid

Variables

Fusion algorithms

Ranked fusion

Relative score fusion

Ranked fusion

Relative score fusion

Weighting & final scores

What can we learn from this?

Additional metadata response

Example

Example with vector specified

Hybrid with a conditional filter

Specify object properties for BM25 search

Oversearch with relativeScoreFusion

BM25 search operator

BM25

Schema configuration

Variables

Additional metadata response

Example query

BM25 with a conditional filter

Search operator

ask

Variables

Example

Additional metadata response

Questions and feedback

Oversearch with `relativeScoreFusion`