Skip to main content

Create objects

The examples on this page demonstrate how to create individual objects in Weaviate.

Use batch import for multiple objects

For creating multiple objects at once, see How-to: Batch Import.

Create an object​

This example creates an object in the JeopardyQuestion collection.

jeopardy = client.collections.get("JeopardyQuestion")

uuid = jeopardy.data.insert({
    "question": "This vector DB is OSS & supports automatic property type inference on import",
    # "answer": "Weaviate",  # properties can be omitted
    "newProperty": 123,  # will be automatically added as a number property
})

print(uuid)  # the return value is the object's UUID
py docs  API docs
Additional information

To create an object, specify the following:

By default, auto-schema creates new collections and adds new properties.

Create an object with a specified vector​

When you create an object, you can provide a vector. (For specifying multiple, named vectors, see below.)

jeopardy = client.collections.get("JeopardyQuestion")
uuid = jeopardy.data.insert(
    properties={
        "question": "This vector DB is OSS and supports automatic property type inference on import",
        "answer": "Weaviate",
    },
    vector=[0.12345] * 1536
)

print(uuid)  # the return value is the object's UUID
py docs  API docs

Create an object with named vectors​

Added in v1.24

When you create an object, you can specify named vectors (if configured in your collection).

reviews = client.collections.get("WineReviewNV")  # This collection must have named vectors configured
uuid = reviews.data.insert(
    properties={
        "title": "A delicious Riesling",
        "review_body": "This wine is a delicious Riesling which pairs well with seafood.",
        "country": "Germany",
    },
    # Specify the named vectors, following the collection definition
    vector={
        "title": [0.12345] * 1536,
        "review_body": [0.31313] * 1536,
        "title_country": [0.05050] * 1536,
    }
)

print(uuid)  # the return value is the object's UUID
py docs  API docs

Create an object with a specified ID​

When you create an object, you can specify an ID.

info

If no ID is provided, Weaviate will generate a random UUID.

properties = {
    "question": "This vector DB is OSS and supports automatic property type inference on import",
    "answer": "Weaviate",
}
jeopardy = client.collections.get("JeopardyQuestion")
uuid = jeopardy.data.insert(
    properties=properties,
    uuid="12345678-e64f-5d94-90db-c8cfa3fc1234"
)

print(uuid)  # the return value is the object's UUID
py docs  API docs

Generate deterministic IDs​

You can generate an ID based on your data object.

info

Object IDs are not randomly generated. The same value always generates the same ID.
Weaviate throws an error if you provide a duplicate ID. Use deterministic IDs to avoid inserting duplicate objects.

from weaviate.util import generate_uuid5  # Generate a deterministic ID

data_object = {
    "question": "This vector DB is OSS and supports automatic property type inference on import",
    "answer": "Weaviate",
}

jeopardy = client.collections.get("JeopardyQuestion")
uuid = jeopardy.data.insert(
    properties=data_object,
    uuid=generate_uuid5(data_object),
)
py docs  API docs
Additional information
To generate deterministic IDs, use one of these methods:

Create an object with cross-references​

Cross-references and query performance

Queries involving cross-references can be slower than queries that do not involve cross-references, especially at scale such as for multiple objects or complex queries.

At the first instance, we strongly encourage you to consider whether you can avoid using cross-references in your data schema. As a scalable AI-native database, Weaviate is well-placed to perform complex queries with vector, keyword and hybrid searches involving filters. You may benefit from rethinking your data schema to avoid cross-references where possible.

For example, instead of creating separate "Author" and "Book" collections with cross-references, consider embedding author information directly in Book objects and using searches and filters to find books by author characteristics.

You can create an object with cross-references to other objects.

questions = client.collections.get("JeopardyQuestion")

questions.data.insert(
    properties=properties,  # A dictionary with the properties of the object
    uuid=obj_uuid,  # The UUID of the object
    references={"hasCategory": category_uuid},  # e.g. {"hasCategory": "583876f3-e293-5b5b-9839-03f455f14575"}
)
py docs  API docs
Additional information

See How-to: Cross-references for more on working with cross-references.

Create an object with geoCoordinates​

Limitations

Currently, geo-coordinate filtering is limited to the nearest 800 results from the source location, which will be further reduced by any other filter conditions and search parameters.

If you plan on a densely populated dataset, consider using another strategy such as geo-hashing into a text datatype, and filtering further, such as with a ContainsAny filter.

If you want to supply a geoCoordinates property, you need to specify the latitude and longitude as floating point decimal degrees:

publications = client.collections.get("Publication")

publications.data.insert(
    properties={
        "headquartersGeoLocation": {
            "latitude": 52.3932696,
            "longitude": 4.8374263
        }
    },
)
py docs  API docs

Validate objects before creation​

Before you create an object, you can validate it against the collection definition.

# Validate is currently not supported with the Weaviate Python client v4
py docs  API docs

Multiple vector embeddings (named vectors)​

Added in v1.24.0

Weaviate collections support multiple named vectors.

Collections can have multiple named vectors.

The vectors in a collection can have their own configurations. Each vector space can set its own index, its own compression algorithm, and its own vectorizer. This means you can use different vectorization models, and apply different distance metrics, to the same object.

To work with named vectors, adjust your queries to specify a target vector for vector search or hybrid search queries.

Questions and feedback​

If you have any questions or feedback, let us know in the user forum.