mirror of
https://github.com/qdrant/landing_page.git
synced 2026-09-25 22:18:30 +02:00
mo feedback
This commit is contained in:
Andrei Vasnetsov
@@ -1,5 +1,5 @@
|
||||
---
|
||||
title: Articles
|
||||
section_title: Our Latest <span>Articles</span>
|
||||
subtitle:
|
||||
subtitle: Check out our latest publications
|
||||
---
|
||||
@@ -1,8 +1,8 @@
|
||||
---
|
||||
title: "We Create The Most Realistic <span>Artificial Intelligence</span>"
|
||||
title: "Make the most of your <span>Unstructured Data</span>"
|
||||
icon:
|
||||
---
|
||||
|
||||
Qdrant is an open-source vector search engine.
|
||||
Qdrant is an vector similarity engine.
|
||||
It deploys as an API service providing search for the nearest high-dimensional vectors.\
|
||||
With Qdrant, embeddings or neural network encoders can be turned into full-fledged applications for matching, searching, recommending, and much more!
|
||||
|
||||
@@ -4,5 +4,5 @@ icon: distribution
|
||||
weight: 50
|
||||
---
|
||||
|
||||
Qdrant is cloud-native and scales horizontally. \
|
||||
No matter how much data you need to serve - Qdrant can always be used with just the right amount of computational resources. (For now - enterprise only)
|
||||
Cloud-native and scales horizontally. \
|
||||
No matter how much data you need to serve - Qdrant can always be used with just the right amount of computational resources. (Currently - Enterprise only)
|
||||
@@ -4,6 +4,5 @@ icon: api
|
||||
weight: 10
|
||||
---
|
||||
|
||||
Qdrant provides the [OpenAPI v3 specification](https://qdrant.github.io/qdrant/redoc/index.html),
|
||||
which allows you to generate a client library in almost any programming language.\
|
||||
Or you can use [ready-made client for Python](https://github.com/qdrant/qdrant_client) or other programming languages with additional functionality.
|
||||
Provides the [OpenAPI v3 specification](https://qdrant.github.io/qdrant/redoc/index.html) to generate a client library in almost any programming language.
|
||||
Alternatively utilise [ready-made client for Python](https://github.com/qdrant/qdrant_client) or other programming languages with additional functionality.
|
||||
@@ -4,5 +4,5 @@ icon: bullseye
|
||||
weight: 20
|
||||
---
|
||||
|
||||
Qdrant implements a unique custom modification of the [HNSW algorithm](https://arxiv.org/abs/1603.09320) for Approximate Nearest Neighbor Search. (TBD) \
|
||||
This algorithm allows Qdrant to search with a [State-of-the-Art speed](https://github.com/erikbern/ann-benchmarks) and apply search filters without [losing the results](https://blog.vasnetsov.com/posts/categorical-hnsw/).
|
||||
Implement a unique custom modification of the [HNSW algorithm](https://arxiv.org/abs/1603.09320) for Approximate Nearest Neighbor Search. (TBD)
|
||||
Search with a [State-of-the-Art speed](https://github.com/erikbern/ann-benchmarks) and apply search filters without [compromising on results](https://blog.vasnetsov.com/posts/categorical-hnsw/).
|
||||
|
||||
@@ -4,6 +4,6 @@ icon: data-1
|
||||
weight: 30
|
||||
---
|
||||
|
||||
Qdrant supports additional payload associated with vectors.
|
||||
Qdrant not only stores payload but also allows filter results based on payload values. \
|
||||
Unlike Elasticsearch k-NN search, Qdrant does not perform post-filtering, so it guarantees that all relevant vectors will be retrieved.
|
||||
Support additional payload associated with vectors.
|
||||
Not only stores payload but also allows filter results based on payload values. \
|
||||
Unlike Elasticsearch post-filtering, Qdrant guarantees all relevant vectors are retrieved.
|
||||
|
||||
@@ -1,9 +1,9 @@
|
||||
---
|
||||
title: Optimized
|
||||
title: Efficient
|
||||
icon: microprocessor
|
||||
weight: 60
|
||||
---
|
||||
|
||||
We made certain Qdrant makes the most of the resources provided.
|
||||
It is developed entirely in Rust language, among other optimizations it implements dynamic query planning and payload data indexing.
|
||||
Effectively utilizes your resources.
|
||||
Developed entirely in Rust language, Qdrant implements dynamic query planning and payload data indexing.
|
||||
Hardware-aware builds are also available for Enterprises.
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
---
|
||||
title: Solutions
|
||||
section_title: Use Cases and Examples For <span>Your Business</span>
|
||||
subtitle: Build your next AI project with Qdrant. Here are just a few examples of applications for which Qdrant can be helpful.
|
||||
section_title: Elevate <span>Your Business</span> with Qdrant
|
||||
subtitle: Here are just a few examples of how Qdrant can help your Business
|
||||
description: Tasks and Problems solved with Qdrant
|
||||
---
|
||||
@@ -20,12 +20,12 @@ short_description: |
|
||||
|
||||
Sometimes text search is not enough.
|
||||
|
||||
Find similar images, detect duplicates, or even find a picture by text description - all of that you can do with Qdrant.
|
||||
Mostly you won't even need to train a neural network for that. Pre-trained models are usually enough to begin with.
|
||||
Qdrant allows you to find similar images, detect duplicates, or even find a picture by text description.
|
||||
No need to train your own neural network, get started with pre-trained models.
|
||||
|
||||
And with Qdrant filters, you can apply arbitrary business logic on top of a similarity search.
|
||||
Qdrant filters enable you to apply arbitrary business logic on top of a similarity search.
|
||||
Look for similar clothes cheaper than $20? Search for a similar artwork published in the last year?
|
||||
Qdrant handles all of these conditions!
|
||||
Qdrant handles all possible conditions!
|
||||
|
||||
For the Demo, we put together a food discovery service - it will show you a lunch based on what you visually like or dislike. It can also search for a place near you. And we didn't train any neural network for that.
|
||||
For the Demo, we put together a food discovery service - it will show you a lunch suggestion based on what you visually like or dislike. It can also search for a place near you.
|
||||
|
||||
|
||||
@@ -18,9 +18,9 @@ short_description: |
|
||||
---
|
||||
|
||||
User behavior can be represented as a semantic vector is similar way as text or images.
|
||||
This vector can represent the user's preferences, behavior patterns, or interest in the product.
|
||||
This vector can represent user preferences, behavior patterns, or interest in the product.
|
||||
|
||||
With Qdrant, user vectors can be updated in real-time, no need to deploy a MapReduce cluster.\
|
||||
Besides, Qdrant will allow you to place arbitrary restrictions on recommendations.
|
||||
What if the user signed up 3 days ago and premium offers are in effect for him?
|
||||
Qdrant will be able to handle this and such conditions.
|
||||
|
||||
|
||||
With Qdrant, user vectors can be updated in real-time, no need to deploy a MapReduce cluster.
|
||||
Understand user behavior in real time.
|
||||
@@ -18,14 +18,15 @@ short_description: |
|
||||
With Qdrant and a pre-trained neural network, you can build and deploy semantic neural search on your data in minutes!
|
||||
---
|
||||
|
||||
In many cases, the usual full-text search does not give the desired result.
|
||||
In many cases, the usual full-text search does not provide the desired result.
|
||||
Documents may have too few keywords, or queries might be too large.
|
||||
In those cases, the search either finds no intersections or returns a lot of irrelevant results.
|
||||
In such cases, the search either finds no intersections or returns a lot of irrelevant results.
|
||||
|
||||
One way to overcome these problems is a neural network-based semantic search.
|
||||
It can be used stand-alone or in conjunction with traditional search.
|
||||
|
||||
The neural search uses **semantic embeddings** instead of keywords and works best with short texts.
|
||||
It can be used stand-alone or in conjunction with traditional search.The neural search uses **semantic embeddings** instead of keywords and works best with short texts.
|
||||
|
||||
With Qdrant and a pre-trained neural network, you can build and deploy semantic neural search on your data in minutes!
|
||||
|
||||
Check out our demo. Compare the results of a semantic search based on a pre-trained transformer NN and a regular full-text search.
|
||||
Check out our demo.
|
||||
Compare the results of a semantic search based on a pre-trained transformer NN and a regular full-text search.
|
||||
|
||||
@@ -10,11 +10,12 @@
|
||||
<!-- Content Column -->
|
||||
<div class="content-column col-lg-6 col-md-12 col-sm-12">
|
||||
<div class="inner-column">
|
||||
<h2><a target="_blank" href="{{ .Site.Params.github }}">Open Source</a></h2>
|
||||
<h1>Neural Search Engine</h1>
|
||||
<div class="sec-title text mb-3">Build your own
|
||||
AI-powered product with Qdrant - <a class="os-link" target="_blank" href="{{ .Site.Params.github }}">open-source</a> vector similarity search engine.
|
||||
<br/><br/>
|
||||
Get started by downloading a pre-build Docker Image - it is only 35Mb!<br>
|
||||
proprietary AI products.
|
||||
<br/>
|
||||
|
||||
|
||||
<div class="code mt-3">
|
||||
<code id="copy-code">docker pull generall/qdrant</code>
|
||||
@@ -24,6 +25,7 @@
|
||||
</button>
|
||||
</span>
|
||||
</div> <br>
|
||||
Get started by downloading a pre-build Docker Image - it is only <b>35Mb!</b><br>
|
||||
|
||||
Take a look at our <a target="_blank" href="{{ .Site.Params.quick_start }}">Quick Start Guide</a>
|
||||
or build your own neural search <br> with our step-by-step <a target="_blank" href="{{ .Site.Params.tutorial }}">Tutorial</a>
|
||||
@@ -190,10 +192,11 @@
|
||||
|
||||
|
||||
{{ with (.Site.GetPage "section" "stack") }}
|
||||
<!--Clients Section-->
|
||||
<!--Clients Section-->
|
||||
<section class="clients-section my-5">
|
||||
<div class="sec-title text-center">
|
||||
<h2>{{ .Title | safeHTML }}</h2>
|
||||
<div class="text">We can integrate with anything, some of our featured technologies</div>
|
||||
</div>
|
||||
<div class="auto-container">
|
||||
<div class="sponsors-outer">
|
||||
@@ -263,7 +266,7 @@
|
||||
<div class="row">
|
||||
<div class="col-12 sec-title text-center">
|
||||
<h2>Get <span>Updates</span> from Qdrant</h2>
|
||||
<div class="text">We will write about new features that appear in Qdrant and novel examples of its applications.
|
||||
<div class="text">We will update you on new features and news regarding Qdrant
|
||||
</div>
|
||||
</div>
|
||||
<div class="content-column offset-lg-3 col-lg-6 col-md-12 col-sm-12">
|
||||
|
||||
Reference in New Issue
Block a user