Embeddings on Qdrant - Vector Search Engine

Aleph Alpha

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Using Aleph Alpha Embeddings with Qdrant

Aleph Alpha is a multimodal and multilingual embeddings’ provider. Their API allows creating the embeddings for text and images, both in the same latent space. They maintain an official Python client that might be installed with pip:

pip install aleph-alpha-client

There is both synchronous and asynchronous client available. Obtaining the embeddings for an image and storing it into Qdrant might be done in the following way:

AWS Bedrock

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Bedrock Embeddings

You can use AWS Bedrock with Qdrant. AWS Bedrock supports multiple embedding model providers.

You’ll need the following information from your AWS account:

Region
Access key ID
Secret key

To configure your credentials, review the following AWS article: How do I create an AWS access key.

With the following code sample, you can generate embeddings using the Titan Embeddings G1 - Text model which produces sentence embeddings of size 1536.

Cohere

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Cohere

Qdrant is compatible with Cohere co.embed API and its official Python SDK that might be installed as any other package:

pip install cohere

The embeddings returned by co.embed API might be used directly in the Qdrant client’s calls:

import cohere
import qdrant_client
from qdrant_client.models import Batch

cohere_client = cohere.Client("<< your_api_key >>")
qdrant_client = qdrant_client.QdrantClient()
qdrant_client.upsert(
 collection_name="MyCollection",
 points=Batch(
 ids=[1],
 vectors=cohere_client.embed(
 model="large",
 texts=["The best vector database"],
 ).embeddings,
 ),
)

If you are interested in seeing an end-to-end project created with co.embed API and Qdrant, please check out the “Question Answering as a Service with Cohere and Qdrant” article.

Fusion Embedding 2

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Fusion Embedding 2

Fusion Embedding 2 is an open-weight multimodal embedding model from Eximius Labs. It maps text, image, video, and audio into a single shared vector space, so content of different types is directly comparable. The model runs on your own hardware and its weights are on Hugging Face.

We’ll look at how to generate Fusion Embedding 2 text vectors and index them in Qdrant with the Python SDK. The lightweight text encoder loads only the base and the trained text head, so it does not pull the audio tower.

Gemini

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Gemini

Google Gemini provides embedding models that are capable of mapping text, image, video, audio, and PDFs and their interleaved combinations thereof into a single, unified vector space. Built on the Gemini architecture, it supports 100+ languages.

The following example shows how to integrate Gemini embeddings with Qdrant:

Setup

# Install the packages from PyPI
# pip install google-genai qdrant-client

// Install the packages from npm
// npm install @google/genai @qdrant/js-client-rest

Let’s see how to use the Embedding Model API to embed documents for retrieval.

Jina Embeddings

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Jina Embeddings

Qdrant is compatible with Jina AI embeddings. You can get a free trial key from Jina Embeddings to get embeddings.

Qdrant users can receive a 10% discount on Jina AI APIs by using the code QDRANT.

Technical Summary

Model	Dimension	Language	MRL (matryoshka)	Context
jina-embeddings-v4	2048 (single-vector), 128 (multi-vector)	Multilingual (30+)	Yes	32768 + Text/Image
jina-clip-v2	1024	Multilingual (100+, focus on 30)	Yes	Text/Image
jina-embeddings-v3	1024	Multilingual (89 languages)	Yes	8192
jina-embeddings-v2-base-en	768	English	No	8192
jina-embeddings-v2-base-de	768	German & English	No	8192
jina-embeddings-v2-base-es	768	Spanish & English	No	8192
jina-embeddings-v2-base-zh	768	Chinese & English	No	8192

Jina recommends using jina-embeddings-v4 for all tasks.

Mistral

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Time: 10 min	Level: Beginner

Mistral

Qdrant is compatible with the new released Mistral Embed and its official Python SDK that can be installed as any other package:

Setup

Install the client

pip install mistralai

And then we set this up:

from mistralai.client import MistralClient
from qdrant_client import QdrantClient
from qdrant_client.models import PointStruct, VectorParams, Distance

collection_name = "example_collection"

MISTRAL_API_KEY = "your_mistral_api_key"
client = QdrantClient(":memory:")
mistral_client = MistralClient(api_key=MISTRAL_API_KEY)
texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

Let’s see how to use the Embedding Model API to embed a document for retrieval.

MixedBread

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Using MixedBread with Qdrant

MixedBread is a unique provider offering embeddings across multiple domains. Their models are versatile for various search tasks when integrated with Qdrant. MixedBread is creating state-of-the-art models and tools that make search smarter, faster, and more relevant. Whether you’re building a next-gen search engine or RAG (Retrieval Augmented Generation) systems, or whether you’re enhancing your existing search solution, they’ve got the ingredients to make it happen.

Mixpeek

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Mixpeek Video Embeddings

Mixpeek’s video processing capabilities allow you to chunk and embed videos, while Qdrant provides efficient storage and retrieval of these embeddings.

Prerequisites

Python 3.7+
Mixpeek API key
Mixpeek client installed (pip install mixpeek)
Qdrant client installed (pip install qdrant-client)

Installation

Install the required packages:

pip install mixpeek qdrant-client

Set up your Mixpeek API key:

from mixpeek import Mixpeek

mixpeek = Mixpeek('your_api_key_here')

Initialize the Qdrant client:

from qdrant_client import QdrantClient

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

Usage

1. Create Qdrant Collection

Make sure to create a Qdrant collection before inserting vectors. You can create a collection with the appropriate vector size (768 for “vuse-generic-v1” model) using:

Nomic

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Nomic

The nomic-embed-text-v1 model is an open source 8192 context length text encoder. While you can find it on the Hugging Face Hub, you may find it easier to obtain them through the Nomic Text Embeddings. Once installed, you can configure it with the official Python client, FastEmbed or through direct HTTP requests.

You can use Nomic embeddings directly in Qdrant client calls. There is a difference in the way the embeddings are obtained for documents and queries.

Nvidia

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Nvidia

Qdrant supports working with Nvidia embeddings.

You can generate an API key to authenticate the requests from the Nvidia Playground.

Setting up the Qdrant client and Nvidia session

import requests
from qdrant_client import QdrantClient

NVIDIA_BASE_URL = "https://ai.api.nvidia.com/v1/retrieval/nvidia/embeddings"

NVIDIA_API_KEY = "<YOUR_API_KEY>"

nvidia_session = requests.Session()

client = QdrantClient(":memory:")

headers = {
 "Authorization": f"Bearer {NVIDIA_API_KEY}",
 "Accept": "application/json",
}

texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

import { QdrantClient } from '@qdrant/js-client-rest';

const NVIDIA_BASE_URL = "https://ai.api.nvidia.com/v1/retrieval/nvidia/embeddings"
const NVIDIA_API_KEY = "<YOUR_API_KEY>"

const client = new QdrantClient({ url: 'http://localhost:6333' });

const headers = {
 "Authorization": "Bearer " + NVIDIA_API_KEY,
 "Accept": "application/json",
 "Content-Type": "application/json"
}

const texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

The following example shows how to embed documents with the embed-qa-4 model that generates sentence embeddings of size 1024.

Ollama

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Using Ollama with Qdrant

Ollama provides specialized embeddings for niche applications. Ollama supports a variety of embedding models, making it possible to build retrieval augmented generation (RAG) applications that combine text prompts with existing documents or other data in specialized areas.

Installation

You can install the required packages using the following pip command:

pip install ollama qdrant-client

Integration Example

The following code assumes Ollama is accessible at port 11434 and Qdrant at port 6333.

OpenAI

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

OpenAI

Qdrant supports working with OpenAI embeddings.

There is an official OpenAI Python package that simplifies obtaining them, and it can be installed with pip:

pip install openai

Setting up the OpenAI and Qdrant clients

import openai
import qdrant_client

openai_client = openai.Client(
 api_key="<YOUR_API_KEY>"
)

client = qdrant_client.QdrantClient(":memory:")

texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

The following example shows how to embed a document with the text-embedding-3-small model that generates sentence embeddings of size 1536. You can find the list of all supported models here.

Prem AI

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Prem AI

PremAI is a unified generative AI development platform for fine-tuning deploying, and monitoring AI models.

Qdrant is compatible with PremAI APIs.

Installing the SDKs

pip install premai qdrant-client

To install the npm package:

npm install @premai/prem-sdk @qdrant/js-client-rest

Import all required packages

from premai import Prem

from qdrant_client import QdrantClient
from qdrant_client.models import Distance, VectorParams

import Prem from '@premai/prem-sdk';
import { QdrantClient } from '@qdrant/js-client-rest';

Define all the constants

We need to define the project ID and the embedding model to use. You can learn more about obtaining these in the PremAI docs.

Snowflake Models

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Snowflake

Qdrant supports working with Snowflake text embedding models. You can find all the available models on HuggingFace.

Setting up the Qdrant and Snowflake models

from qdrant_client import QdrantClient
from fastembed import TextEmbedding

qclient = QdrantClient(":memory:")
embedding_model = TextEmbedding("snowflake/snowflake-arctic-embed-s")

texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

import {QdrantClient} from '@qdrant/js-client-rest';
import { pipeline } from '@xenova/transformers';

const client = new QdrantClient({ url: 'http://localhost:6333' });

const extractor = await pipeline('feature-extraction', 'Snowflake/snowflake-arctic-embed-s');

const texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

The following example shows how to embed documents with the snowflake-arctic-embed-s model that generates sentence embeddings of size 384.

Superlinked

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Superlinked

Superlinked is a self-hosted inference engine (SIE) that serves 85+ embedding models (dense, sparse, and multivector / ColBERT) from a single endpoint. The sie-qdrant package lets you use SIE as the embedding provider for Qdrant collections. SIE encodes your text into vectors, and you store and search them in Qdrant.

sie-qdrant is currently Python only. TypeScript support is not yet available.

Installation

pip install sie-qdrant

This installs sie-sdk and qdrant-client (v1.7+) as dependencies. You also need a running SIE instance; see the Superlinked quickstart for deployment options (Docker, GPU).

Twelve Labs

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Twelve Labs

Twelve Labs Embed API provides powerful embeddings that represent videos, texts, images, and audio in a unified vector space. This space enables any-to-any searches across different types of content.

By natively processing all modalities, it captures interactions like visual expressions, speech, and context, enabling advanced applications such as sentiment analysis, anomaly detection, and recommendation systems with precision and efficiency.

We’ll look at how to work with Twelve Labs embeddings in Qdrant via the Python and Node SDKs.

Upstage

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Upstage

Qdrant supports working with the Solar Embeddings API from Upstage.

Solar Embeddings API features dual models for user queries and document embedding, within a unified vector space, designed for performant text processing.

You can generate an API key to authenticate the requests from the Upstage Console.

Setting up the Qdrant client and Upstage session

import requests
from qdrant_client import QdrantClient

UPSTAGE_BASE_URL = "https://api.upstage.ai/v1/solar/embeddings"

UPSTAGE_API_KEY = "<YOUR_API_KEY>"

upstage_session = requests.Session()

client = QdrantClient(url="http://localhost:6333")

headers = {
 "Authorization": f"Bearer {UPSTAGE_API_KEY}",
 "Accept": "application/json",
}

texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

import { QdrantClient } from '@qdrant/js-client-rest';

const UPSTAGE_BASE_URL = "https://api.upstage.ai/v1/solar/embeddings"
const UPSTAGE_API_KEY = "<YOUR_API_KEY>"

const client = new QdrantClient({ url: 'http://localhost:6333' });

const headers = {
 "Authorization": "Bearer " + UPSTAGE_API_KEY,
 "Accept": "application/json",
 "Content-Type": "application/json"
}

const texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

The following example shows how to embed documents with the recommended solar-embedding-1-large-passage and solar-embedding-1-large-query models that generates sentence embeddings of size 4096.

Voyage AI

info@qdrant.tech (Andrey Vasnetsov) — Mon, 01 Jan 0001 00:00:00 +0000

Voyage AI

Qdrant supports working with Voyage AI embeddings. The supported models’ list can be found here.

You can generate an API key from the Voyage AI dashboard to authenticate the requests.

Setting up the Qdrant and Voyage clients

from qdrant_client import QdrantClient
import voyageai

VOYAGE_API_KEY = "<YOUR_VOYAGEAI_API_KEY>"

qclient = QdrantClient(":memory:")
vclient = voyageai.Client(api_key=VOYAGE_API_KEY)

texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

import {QdrantClient} from '@qdrant/js-client-rest';

const VOYAGEAI_BASE_URL = "https://api.voyageai.com/v1/embeddings"
const VOYAGEAI_API_KEY = "<YOUR_VOYAGEAI_API_KEY>"

const client = new QdrantClient({ url: 'http://localhost:6333' });

const headers = {
 "Authorization": "Bearer " + VOYAGEAI_API_KEY,
 "Content-Type": "application/json"
}

const texts = [
 "Qdrant is the best vector search engine!",
 "Loved by Enterprises and everyone building for low latency, high performance, and scale.",
]

The following example shows how to embed documents with the voyage-large-2 model that generates sentence embeddings of size 1536.