- 113
- 152 355
Qdrant - Vector Database & Search Engine
Germany
Приєднався 21 чер 2022
Open-source vector database and semantic search engine.
Optimizing Document Retrieval with ColPali and Qdrant's Binary Quantization
ColPali is a new document retrieval model and training strategy that leverages Vision Language Models (VLMs) to index document pages based on their visual elements. In this video, we'll demonstrate how ColPali works and how you can use Binary Quantization to achieve faster and more efficient results without compromising accuracy.
ColPali paper: arxiv.org/abs/2407.01449
ColPali and Qdrant by Daniel van Strien: danielvanstrien.xyz/posts/post-with-code/colpali-qdrant/2024-10-02_using_colpali_with_qdrant.html
Notebook: colab.research.google.com/drive/14FqLc0N_z92_VgL_zygWV5pJZkaskyk7?usp=sharing
Qdrant's Discord Community: discord.com/invite/qdrant
ColPali paper: arxiv.org/abs/2407.01449
ColPali and Qdrant by Daniel van Strien: danielvanstrien.xyz/posts/post-with-code/colpali-qdrant/2024-10-02_using_colpali_with_qdrant.html
Notebook: colab.research.google.com/drive/14FqLc0N_z92_VgL_zygWV5pJZkaskyk7?usp=sharing
Qdrant's Discord Community: discord.com/invite/qdrant
Переглядів: 1 719
Відео
Visualizing Vector Embeddings: Qdrant’s WebUI Graph Tool
Переглядів 928Місяць тому
Access the dashboard by creating a free account: cloud.qdrant.io Quickstart: qdrant.tech/documentation/quickstart-cloud/
RAG Movie Recommendations chatbot with Qdrant and n8n
Переглядів 2 тис.Місяць тому
Ever wanted to watch a vampire movie but not something overly romantic like Twilight? Or a magical adventure, but you’re done rewatching Harry Potter for the millionth time? Well, then we won’t just give you a fish; we'll hand you the fishing rod!:) Learn how to build your own RAG movie recommendation chatbot - to get exactly what you like and avoid what you want to avoid - using the free-tier ...
What is RAG? Building Better LLM Systems with Qdrant
Переглядів 1,1 тис.3 місяці тому
Large Language Models (LLMs) often struggle to keep up with new information. So, how can we fix this? In this video, discover how Retrieval-Augmented Generation (RAG) integrates real-time data from vector databases like Qdrant to ensure accurate and relevant answers instantly. Join us as we explore fundamental RAG concepts, including indexing, chunking, embedding, retrieving, and generating dat...
How Vector Search Algorithms Work: An Intro to Qdrant
Переглядів 3,5 тис.4 місяці тому
Learn how Qdrant analyzes context and meaning in data to deliver precise recommendations and search results. See how it transforms unstructured data like text, images, audio, and video into high-dimensional vectors for lightning-fast, semantically relevant matches. Join the Qdrant Vector Search community: discord.gg/qdrant Check out some of our other videos: Optimize RAG Resource Use With Seman...
Recommendation system with Qdrant and sparse vectors (Collaborative Filtering)
Переглядів 1 тис.4 місяці тому
Let's build a movie recommendation system using Qdrant and sparse vectors. In this video, let's see how it's possible to implement collaborative filtering with Qdrant. Links: - Sparse vectors: qdrant.tech/articles/sparse-vectors/ - Demo: github.com/infoslack/qdrant-example/blob/main/sparse-vectors/collaborative-filtering.ipynb
Advanced RAG - Self Querying Retrieval
Переглядів 6 тис.5 місяців тому
Semantic search is powerful, but it's not the answer to everything. For example, you don't need a semantic search if you're searching for a year or a specific name. Instead, use direct lookups for those cases. A semantic search should be used to extract meaning from text, like when looking for a movie and specifying the year or searching for music on Spotify by artist name. In this video, we'll...
Optimize RAG Resource Use With Semantic Cache
Переглядів 5 тис.6 місяців тому
Optimize RAG Resource Use With Semantic Cache
Deploy a Production-Ready Vector Database in 5 Minutes With Qdrant Hybrid Cloud
Переглядів 1,6 тис.6 місяців тому
Deploy a Production-Ready Vector Database in 5 Minutes With Qdrant Hybrid Cloud
Building Search/RAG for an OpenAPI spec - Nick Khami | Vector Space Talk #022
Переглядів 5567 місяців тому
Building Search/RAG for an OpenAPI spec - Nick Khami | Vector Space Talk #022
Advancements and Challenges in RAG Systems - Syed Asad | Vector Space Talks #021
Переглядів 3917 місяців тому
Advancements and Challenges in RAG Systems - Syed Asad | Vector Space Talks #021
Gen AI and Vector Search - Iveta Lohovska | Vector Space Talks #020
Переглядів 2407 місяців тому
Gen AI and Vector Search - Iveta Lohovska | Vector Space Talks #020
Teaching Vector Databases at Scale - Alfredo Deza | Vector Space Talks #019
Переглядів 2177 місяців тому
Teaching Vector Databases at Scale - Alfredo Deza | Vector Space Talks #019
Exploring Qdrant concepts - Collections
Переглядів 1,2 тис.7 місяців тому
Exploring Qdrant concepts - Collections
How to meow on the long tail with Cheshire Cat AI? - Piero and Nicola | Vector Space Talks #018
Переглядів 2498 місяців тому
How to meow on the long tail with Cheshire Cat AI? - Piero and Nicola | Vector Space Talks #018
Chatbot with RAG, using LangChain, OpenAI, and Groq
Переглядів 23 тис.8 місяців тому
Chatbot with RAG, using LangChain, OpenAI, and Groq
Best RAG to unleash the power of AI - Guillaume Marquis of VirtualBrain | Vector Space Talks #017
Переглядів 2588 місяців тому
Best RAG to unleash the power of AI - Guillaume Marquis of VirtualBrain | Vector Space Talks #017
Talk with YouTube without paying a cent - Francesco Saverio Zuppichini | Vector Space Talks #016
Переглядів 5598 місяців тому
Talk with UA-cam without paying a cent - Francesco Saverio Zuppichini | Vector Space Talks #016
Production-scale RAG for Real-Time News Distillation - Robert Caulk | Vector Space Talks #015
Переглядів 8098 місяців тому
Production-scale RAG for Real-Time News Distillation - Robert Caulk | Vector Space Talks #015
Getting started with DSPy tutorial
Переглядів 18 тис.8 місяців тому
Getting started with DSPy tutorial
Embeddings, Similarity Search and Qdrant
Переглядів 2,9 тис.9 місяців тому
Embeddings, Similarity Search and Qdrant
Insight Generation Platform for LifeScience Corporation - Hooman Sedghamiz | Vector Space Talks #014
Переглядів 2479 місяців тому
Insight Generation Platform for LifeScience Corporation - Hooman Sedghamiz | Vector Space Talks #014
The challenges in using LLM-as-a-Judge - Sourabh Agrawal | Vector Space Talk #013
Переглядів 9749 місяців тому
The challenges in using LLM-as-a-Judge - Sourabh Agrawal | Vector Space Talk #013
Vector Search for Content-Based Video Recommendation - Gladys and Sam | Vector Space Talk #012
Переглядів 3139 місяців тому
Vector Search for Content-Based Video Recommendation - Gladys and Sam | Vector Space Talk #012
Retrieval in Generative Language Model Workflows - Mikko Lehtimäki | Vector Space Talk #011
Переглядів 3659 місяців тому
Retrieval in Generative Language Model Workflows - Mikko Lehtimäki | Vector Space Talk #011
Qdrant x Dust: How Vector Search helps make work work better - Stan Polu | Vector Space Talk #010
Переглядів 3389 місяців тому
Qdrant x Dust: How Vector Search helps make work work better - Stan Polu | Vector Space Talk #010
Qdrant Team Las Palmas Off-site 2023 Recap!
Переглядів 36510 місяців тому
Qdrant Team Las Palmas Off-site 2023 Recap!
Indexify Unveiled - Diptanu Gon Choudhury | Vector Space Talk #009
Переглядів 42310 місяців тому
Indexify Unveiled - Diptanu Gon Choudhury | Vector Space Talk #009
How to make retrieval time intelligent meaning retrive the latest without using metadata/payload. Latest can be from any year going back to previous years that is latest available.
Colpali (and other similar models) are really game changing. Manually handcraft pdfs with text images table etc is really a pain in the ass!
Want video on how embeddings work
very insightful, I learned a lot!
Where is the next tutorial?
Thanks for this video. Unfortunately, you lost me at 28:00, where the main.py script ran into a lot of connection errors. Also, where does qdrant come from? There is no service running at :6333 on my machine.
You’re the absolute best! 💐
Very interesting project !
Nice!
Great video! Could you please consider increase the audio gain when recording the video? Thanks!
I always enjoy talking to “rob”
I do have a question about it: What if I was to add another LLM layer after the cache returnable? To kind of change the answer, to do not feel that we are always getting the same answer. What are the downsides of doing that, and how could I do it properly?
Hello, Evgenia, great job!!! I'm from Brazil, so please forgive my English. I replicated your project and need to make some changes. I need to recommend products to customers, so I select a customer (or even several customers) from a drop-down list, then select some filters such as: region, average purchase, spending and others, then the automation analyzes the customer's purchase history and recommends the products. Any suggestions on how I can do this? I would be immensely grateful. Bye
Jenkins Way
Rodrick Plaza
thanks for creating this video, very helpful to me, please continuous create another usecase using this n8n workflow thanks
Absolutely!:) Planning a webinar on it with n8n in December:)
Maxine Forest
Monserrat Spur
Camille Parkway
Sheridan Cliff
Buckridge Oval
Predovic Forge
Seth Ford
Conner Knolls
Hackett Manors
Thanks for the explanation broww
Evgenia, great video. If you have free time, shoot more videos using n8n and Qdrant. It is very interesting.
Glad you liked it!:) Will think about use cases!
Jerde Path
Nice. Thanks
Great video! I'm curious: are there any tools or strategies that could help determine when to apply filters to embeddings based on user input? For example, identifying when to filter results by price. With Qdrant, you could leverage payloads for this purpose... This would allow semantic search on a filtered set.
ZCMNBV❤️🔥😊🏕🌄🚅😴🇧🇭
Even though the content is amazingly excellent the accent is poor and it makes the content pretty hard to understand
Hey Evgeniya! I followed the rabbit trail from the N8N community meetup video to this one, which was fantastic. Thanks for doing this. I am strategically all in on the whole low-code automation and RAG bandwagon, and have been doing copious research. I loved this presentation at the community meetup, as well as following along with this. I'd like to make the right decision for our organization when it comes to handling embeddings and data retrieval needs for our particular offerings. I have a few questions. I was hoping you could point me in the right direction. Would the comments here be the best place to do that, or is there an email address or chat platform you'd point me towards? I'd like to give you an idea of where we're at, what our goals are, what I know, and what I have access to in a general way, but as specifically as needs be, so you can tell me why or why not I should look at a solution such as yours. I have several books and things like that I'm reading, and a few different choices that I'm weighing. I want to make sure something fits within my personal skill set, as well as the outcomes that I'm looking to specifically produce. I am, however, willing to learn. It all comes down to the return on time investment, I guess. Thanks for replying to this. I look forward to hearing from you more. 😊
I signed up for the free cluster today, you can probably find my email if that works for you.
Awesome video <3 Could you share link of Google Colab? Thanks bro <3
how to use the existing ddrant database as reteriver
Poor edit
Can you share the Google Colab link?
Awesome Webinar. Keep going freens
Hi , what is the "Qdrant Url " and where can I get it ?
i was amazing with your speaking speed lol
Beahan Mall
Bailee Ramp
O'Reilly Ports
Schneider Ports
Mateo Cliff
Kyra River
Reynolds Drive
Mohamed Plains
Sally Plaza
Helene Shoal