landing_page/qdrant-landing/content/documentation/quick-start.md

title, weight, aliases

title	weight	aliases
Quickstart	11	quick_start

Quickstart

Recommended Workflow

There are two ways of using Qdrant locally: 1) Local Mode and 2) Docker Mode. If you are a Python developer, we recommend that you first try Local Mode in Qdrant Client, as it only takes a few moments to get setup. Then, you may further experiment with Qdrant Docker containers. When you are more comfortable with Qdrant, then you should deploy your app to a Free Tier Qdrant Cloud cluster.

Local Mode	Docker Mode	Qdrant Cloud

Option 1: Local Mode

Prerequisite: First make sure you have the latest version of Python installed.

Install Qdrant

pip install qdrant-client

Initialize Qdrant Client

from qdrant_client import QdrantClient

client = QdrantClient(":memory:")
# or
client = QdrantClient(path="path/to/db")  # Persists changes to disk

Option 2: Docker Mode

Prerequisite: First make sure Docker is installed and running on your system.

Download the image

docker pull qdrant/qdrant

Run the service

docker run -p 6333:6333 \
    -v $(pwd)/qdrant_storage:/qdrant/storage:z \
    qdrant/qdrant

Under the default configuration all data will be stored in the ./qdrant_storage directory. This will also be the only directory that both the Container and the host machine can both see.

Qdrant should now be accessible at localhost:6333

Initialize Qdrant Client

from qdrant_client import QdrantClient

client = QdrantClient("localhost", port=6333)

By default, Qdrant starts with encryption or authentication . This means anyone with network access to your machine can access your Qdrant container instance. Please read Security carefully for details on how to secure your instance.

Running vector search queries

In this short example, you will use the Python client to create a Qdrant collection, load data into it and run a basic search query.

Step 1. Create a collection

You will be storing all of your vector data in a Qdrant collection. Let's call it test_collection. This collection will be using a doc product distance metric to compare vectors.

from qdrant_client.http.models import Distance, VectorParams

client.recreate_collection(
    collection_name="test_collection",
    vectors_config=VectorParams(size=4, distance=Distance.DOT),
)

Response:

True

Step 2: Add vectors

Let's now add a few vectors with a payload. Payloads are other data you want to associate with the vector:

from qdrant_client.http.models import PointStruct

operation_info = client.upsert(
    collection_name="test_collection",
    wait=True,
    points=[
        PointStruct(id=1, vector=[0.05, 0.61, 0.76, 0.74], payload={"city": "Berlin"}),
        PointStruct(id=2, vector=[0.19, 0.81, 0.75, 0.11], payload={"city": ["Berlin", "London"]}),
        PointStruct(id=3, vector=[0.36, 0.55, 0.47, 0.94], payload={"city": ["Berlin", "Moscow"]}),
        PointStruct(id=4, vector=[0.18, 0.01, 0.85, 0.80], payload={"city": ["London", "Moscow"]}),
        PointStruct(id=5, vector=[0.24, 0.18, 0.22, 0.44], payload={"count": [0]}),
        PointStruct(id=6, vector=[0.35, 0.08, 0.11, 0.44]),
    ]
)
print(operation_info)

Response:

operation_id=0 status=<UpdateStatus.COMPLETED: 'completed'>

Step 3: Run a query

Let's ask a basic question - Which of our stored vectors are most similar to the query vector [0.2, 0.1, 0.9, 0.7]?

search_result = client.search(
    collection_name="test_collection",
    query_vector=[0.2, 0.1, 0.9, 0.7], 
    limit=3
)
print(search_result)

Response:

ScoredPoint(id=4, version=0, score=1.362000031210482, payload={'city': ['London', 'Moscow']}, vector=None), 
ScoredPoint(id=1, version=0, score=1.2729999996721744, payload={'city': 'Berlin'}, vector=None), 
ScoredPoint(id=3, version=0, score=1.2080000013113021, payload={'city': ['Berlin', 'Moscow']}, vector=None)

The results are returned in decreasing si,ilarity order. Note that payload and vector data is missing in these results by default. See payload and vector in the result on how to enable it.

Step 4: Add a filter

We can narrow down the results further by filtering by payload. Let's find the closest results that include "London".

from qdrant_client.http.models import Filter, FieldCondition, MatchValue


search_result = client.search(
    collection_name="test_collection",
    query_vector=[0.2, 0.1, 0.9, 0.7], 
    query_filter=Filter(
        must=[
            FieldCondition(
                key="city",
                match=MatchValue(value="London")
            )
        ]
    ),
    limit=3
)
print(search_result)

Response:

ScoredPoint(id=4, version=0, score=1.362000031210482, payload={'city': ['London', 'Moscow']}, vector=None), 
ScoredPoint(id=2, version=0, score=0.8709999993443489, payload={'city': ['Berlin', 'London']}, vector=None)

You have just conducted vector search. You loaded vectors into a database and queried the database with a vector of your own. Qdrant found the closest results and presented you with a similarity score.

Next Steps

Now you know how Qdrant works. Getting started with Qdrant Cloud is just as easy. Create an account and use our SaaS completely free. We will take care of infrastructure maintenance and software updates.

To move onto some more complex examples of vector search, read our Tutorials and create your own app with the help of our Examples.