Weaviate • vector database @weaviate_io Twitter profile | Pikagi

Pikagi

Weaviate • vector database

@weaviate_io

11,681

Followers

3,223

Following

751

Media

3,503

Statuses

The easiest way to build and scale AI applications. 🐙 📰

🌎

https://t.co/yIzs0MXLU0

Joined May 2018

Don't wanna be here? Send us removal request.

Pinned Tweet

@weaviate_io

Weaviate • vector database

3 months

🐍 The #Python client for @weaviate_io just got a major upgrade! v4 is out of beta, and brings you faster performance with gRPC, enhanced IDE support, and next-level type safety. It also comes with many quality-of-life improvements like import rate limits, iterators and more!…

Tweet media one

2

9

27

Last Seen Profiles

@mattRlive

@wonpilaroid

@yukinik4

@world_gummy

@OchoReacts

@reddskyy_

@NagasakiPengin

@spiritofniter

@tkaiser_science

@SECNigeria

@Gerhardhormann

@sammisilber

@NAYouthCenter

@Alex_Danco

@Ldi_32

@Futbolsura

@TyStinson4

@menkosaurus

@drjackUK

@rrriinv

@JeanCarlosCCC

@Hora25

@gonbe0211

@tomhendo2

@architect_eth

@heroine_fine

@victorgodinho

@Annin76

@lovinqmoon_

@Unlimitedv59

@hugosalazar_

@poshatami

@wael_alyzedi

@CoachJD561

@marykatelillis

@toretore__oripa

@weaviate_io

Weaviate • vector database

1 year

🔥 We are excited to announce our new generative search module 🔎 The generative @OpenAI module pipes your Weaviate results directly through the GPT-3 model! 🤩 After vector search and hybrid search, we now have generative search! 💥 Check out the 🧵!

5

51

232

@weaviate_io

Weaviate • vector database

1 year

⚡️ #AutoGPT harnesses the power of GPT-4 and GPT-3.5 🔗 It chains together "thoughts" and completes various tasks autonomously 🧠 Learn how you can use it with #Weaviate to enable a short-term and long-term memory ✍️ By @ecardenas300 and @_jphwang :

Tweet card media

Giving Auto-GPT Long-Term Memory with Weaviate | Weaviate - Vector Database

Learn about Auto-GPT and how to give it long-term memory with Weaviate!

5

34

165

@weaviate_io

Weaviate • vector database

1 year

🥳 Super proud and happy to announce that we've closed our Series-B led by @IndexVentures with participation from @BatteryVentures and our existing investors ( @espricewright , @Dthakker02 , and @DanelDayan . Welcome to Weaviate)! 📰 Coverage in The Information:…

Tweet media one

16

39

161

@weaviate_io

Weaviate • vector database

6 months

Verba, our open-source RAG framework 🐕 Keep your data private and 100% free using open-source models like SentenceTransformers and run Weaviate locally. Repo: Learn more about the modular architecture that powers you to build in the thread!

Tweet media one

2

32

132

@weaviate_io

Weaviate • vector database

17 days

We've aggregated all of the DSPy resources from the Weaviate team on one page! It is broken into two categories: 1. Hands on Learning and 2. Read and Listen 📚 DSPy round-up: We're also very excited to host our in-person meetup with DSPy, @arizeai , and…

Tweet media one

3

29

132

@weaviate_io

Weaviate • vector database

10 months

You can build a chatbot using the new open source Llama 2 model with these tools 🦙 • @replicatehq for the endpoint to the llama13b-v2-chat model • @llama_index as the framework • @weaviate_io for the vector store Notebook:

Tweet media one

1

27

119

@weaviate_io

Weaviate • vector database

1 year

🧠 🔌 With #ChatGPT 's Retrieval Plugin powered by #Weaviate 's Vector Database, you can now conduct generative searches using your own data and overcome ChatGPT's lack of long-term memory! 🚀 ✍️ Read the blog post by @ZainHasan6 :

6

15

114

@weaviate_io

Weaviate • vector database

6 months

Using LLMs to evaluate RAG systems lets us build better applications faster! 🧫 We have a new blog post from @ecardenas300 and @CShorten30 presenting new evaluation tools such as Ragas and ARES, and the current state of the art in RAG:

Tweet media one

12

32

112

@weaviate_io

Weaviate • vector database

1 month

Ready to build a “Chat with your code” application with @ollama , Weaviate, and @llama_index on @LightningAI Studio? We’ve just published a Lightning AI Studio template that you can simply copy. You’ll be up and running with your Retrieval-Augmented Generation (RAG) pipeline in…

5

45

108

@weaviate_io

Weaviate • vector database

2 months

Generative Feedback Loops will transform how the world makes use of its data. Introducing Hurricane 🌀, an open-source web application that converts questions or topics into blog posts! We then feed these blog posts back into Weaviate to create an ever-growing knowledge base of…

Tweet media one

3

26

96

@weaviate_io

Weaviate • vector database

1 year

😲 Building a custom chatbot using @LangChainAI and #Weaviate can be done in 20 lines of code 🖇 Simply connect to your Weaviate instance and specify the class and properties that you want LangChain to use (🥁) Last but not least, specify the @OpenAI model

Tweet media one

2

27

92

@weaviate_io

Weaviate • vector database

1 month

⭐ Updated @LangChainAI integration ⭐ We completely rewrote the Weaviate LangChain integration using the Python v4 client! It now supports all Weaviate features, including multitenancy and autocut. Also, hybrid search is even easier to implement! Docs:

Tweet media one

3

26

90

@weaviate_io

Weaviate • vector database

1 year

🤩 Learn how to combine LangChain and Weaviate to build a custom LLM chatbot powered with semantic search by @ecardenas300 ⚙️ LangChain offers various implementations to solve hallucination and limited token lengths! Read this blog post 👇

Tweet card media

Combining LangChain and Weaviate | Weaviate - Vector Database

LangChain is one of the most exciting new tools in AI. It helps overcome many limitations of LLMs, such as hallucination and limited input lengths.

4

21

79

@weaviate_io

Weaviate • vector database

3 months

Reranking is one of quickest ways to get more accurate search results! You can take this one step further by fine-tuning the reranker 🤝 In this blog post, @ecardenas300 goes over how you can fine-tune @cohere 's reranker model and generate synthetic data with DSPy:…

Tweet media one

2

13

78

@weaviate_io

Weaviate • vector database

2 months

DSPy is a framework for programming LLM applications. You can now build next-level applications using Weaviate and DSPy! Watch the hands-on video by @ecardenas300 where she takes questions and converts them into entire blog posts. Video: Demo:…

Tweet media one

0

16

70

@weaviate_io

Weaviate • vector database

1 year

Exciting news! Weaviate now supports the brand-new Google Pathways Language Model (PaLM) through two new modules. Enhance your #VectorSearch and generative tasks using PaLM's 540-billion 🤯 parameters!

Tweet media one

3

13

62

@weaviate_io

Weaviate • vector database

2 months

Are you ready to take your naive Retrieval Augmented Generation (RAG) pipelines to the next level? Try applying a few advanced RAG techniques with @llama_index and Weaviate. In her latest article, @helloiamleonie discusses current advanced RAG techniques, which can be…

Tweet media one

0

14

62

@weaviate_io

Weaviate • vector database

8 months

Thanks to RAG, Customer Relationship Management (CRM) systems are getting a boost 🤝 @dlthub built a demo using Verba to chat with Zendesk tickets in just a few steps. Blog post: Demo:

Tweet media one

1

11

59

@weaviate_io

Weaviate • vector database

1 year

🥳 New release - Weaviate 1.18 is out now! 🚀 With a set of new great features, performance and UX improvements, and fixes Read the blog post: Release notes: Release podcast: 🧵 All features of v.1.18

5

14

50

@weaviate_io

Weaviate • vector database

2 months

DSPy simplifies the process of building LLM applications by prioritizing programming over prompting. With DSPy and Weaviate, you can create next-generation LLM applications more efficiently! To learn more, check out the resources shared by our colleagues: Gentle Introduction to…

Tweet media one

0

26

51

@weaviate_io

Weaviate • vector database

25 days

Did you know about the Binoculars technique to reliably tell apart fake, LLM-generated, text from human-written text? Or have you heard about Modular RAG? What about Matryoshka Embeddings? These are just a few posts in our new paper reviews page, where we create 1-2 minute…

3

8

47

@weaviate_io

Weaviate • vector database

1 month

It's kind of crazy to think that you can throw out 97% of the information in a vector embedding and still retrieve accurately. But it's true! In this blog, we provide a super simple explanation of the buzz around binary quantization and help you get started with implementing and…

Tweet media one

3

10

48

@weaviate_io

Weaviate • vector database

1 year

💡 As a vector DB company, we rely on the powerful ML models our partners create. 👉 @CohereAI is an example of such a partner; their multilingual model is just amazing! 💪 The power of Weaviate + Cohere is shown in this awesome article.

The Embedding Archives: Millions of Wikipedia Article Embeddings in Many Languages

There’s no denying that we’re in the midst of a revolutionary time for Language AI. Developers are waking up to the vast emerging capabilities of language understanding and generation models. One of...

1

15

47

@weaviate_io

Weaviate • vector database

25 days

Prompt engineering is a key part of getting better responses from Large Language Models. The particular wording of the instructions and examples can determine the fate of the performance 😵‍💫 DSPy Compilers simplify the process of writing the perfect prompt! In this blog post,…

Tweet media one

5

10

42

@weaviate_io

Weaviate • vector database

1 year

🥱 Say goodbye to googling for Weaviate docs 🔎 now has SITE SEARCH! 🚀 Quickly search through our docs without leaving the website Happy searching! #SiteSearch #WeaviateDocs

3

10

42

@weaviate_io

Weaviate • vector database

10 months

Weaviate 1.20 is here! 🚢 Covering: • Multi-tenancy • PQ + re-scoring • Autocut • Search re-ranking • New hybrid search ranking algorithm Learn more about each feature! Blog post: Release podcast:

6

10

41

@weaviate_io

Weaviate • vector database

9 months

One distance metric does not fit all embedding models 📏 For example, the @cohere multilingual model uses dot product, whereas the English model uses cosine. Read this blog post by @ecardenas300 where she covers the various distance metrics in Weaviate:

Tweet media one

5

15

41

@weaviate_io

Weaviate • vector database

2 years

🎉 Weaviate v1.10.0 is here! Besides a lot of new features, we've released the OpenAI Weaviate module that directly integrates with @OpenAI 's embeddings API! You can now vectorize, search through, and even mix different ML models with the power of GPT-3 all out-of-the-box!

Tweet media one

4

16

42

@weaviate_io

Weaviate • vector database

9 days

Cognee allows you to introduce more predictability and management into AI workflows through graph architectures, vector stores, and automation. With their new release, they’ve added more awesome features and integrations, like DSPy graph generation and @neo4j support! GitHub:…

Tweet media one

0

15

43

@weaviate_io

Weaviate • vector database

1 year

🎉 Weaviate 1.19 is now available! 🚀 Packed with new features, improved performance, and UX enhancements. 📚 Learn more about this release by checking out our: 📝 Blog post: 📑 Release notes:

1

6

40

@weaviate_io

Weaviate • vector database

3 months

HNSW is a fast, memory-efficient algorithm for finding similar data points in large datasets. It uses a multi-layered graph to 'jump' over unrelated data, reducing search time and memory usage. Here are five key terms and concepts to help you better understand HNSW 🧵 Read more…

Tweet media one

1

5

40

@weaviate_io

Weaviate • vector database

11 months

LLMs and search go hand in hand 🤝 In this blog post, @CShorten30 and @ecardenas300 share five components on the intersection between LLMs and search! Read the blog post:

Tweet card media

Large Language Models and Search | Weaviate - Vector Database

Learn about the intersection between LLMs and Search

1

10

36

@weaviate_io

Weaviate • vector database

7 months

How can a workation in Croatia enhance team collaboration and infuse positive energy into your team's dynamics? Our recent gathering in Croatia worked wonders, forging strong bonds, boosting teamwork, and creating memorable memories. 🏝️

5

9

37

@weaviate_io

Weaviate • vector database

6 months

We're teaming up with @DeepLearningAI for a short course: Vector Databases - From Embeddings to Applications 🚀 Ready to get started with vector databases? Enroll now:

1

13

36

@weaviate_io

Weaviate • vector database

2 months

Want to get some hands-on experience with building Retrieval-Augmented Generation (RAG) pipelines? Why not join the current @kaggle competition “Google – AI Assistants for Data Tasks with Gemma”? To get started, you can check out @helloiamleonie 's Kaggle Notebooks: 📌 RAG with…

Tweet media one

0

11

36

@weaviate_io

Weaviate • vector database

4 months

Verba, our Retrieval Augmented Generation demo, uses hybrid search 👀 Hybrid search can help a RAG app retrieve the most relevant results for answer generation because it has the best of both worlds - the combination of keyword and vector search. This allows Verba to retrieve…

Tweet media one

2

13

35

@weaviate_io

Weaviate • vector database

1 year

🙋 Hello, Weaviate friends! 🚀 Join our (first!) introduction to Weaviate workshop with @_jphwang on April 4th, 1-2 PM CET 📚 Covering vector databases + language models, and hands-on demos using Python ✍️ Limited spaces available - sign up now at

Tweet media one

1

10

33

@weaviate_io

Weaviate • vector database

1 year

🎉 We’re happy to announce a new #Weaviate module – text2vec-cohere – which allows you to directly integrate with @CohereAI 's Large Language Models (LLMs) 👉

1

12

34

@weaviate_io

Weaviate • vector database

11 months

Retrieval-Augmented Generation enables you to combine large language models with your data 🚢 @llama_index and Weaviate are a powerful RAG stack for your LLM application @jerryjliu0 shares an overview on LlamaIndex and goes through a demo notebook

Tweet card media

LlamaIndex and Weaviate | Weaviate - Vector Database

An introductory overview of LlamaIndex, the LLM framework

2

10

34

@weaviate_io

Weaviate • vector database

10 months

Weaviate Turns Four! 🎂🚀 Co-founders @etiennedi & @bobvanluijt reflect on an incredible journey of growth, paradigm shifts, and an unwavering belief in AI. Read their story and learn about the changes that shaped Weaviate's path:

10

6

32

@weaviate_io

Weaviate • vector database

8 months

At Weaviate, we 💚 open source, so today, we’re releasing a full-stack open source RAG application. Meet Verba, The Golden RAGtriever! 🐕 Verba is designed to offer a streamlined, user-friendly interface for RAG apps! Video:

1

10

33

@weaviate_io

Weaviate • vector database

8 months

Don't know where to start learning about vector databases? Vector databases can help enhance your AI applications by vectorizing data into embedding spaces that allow for quick retrieval and compariso 🤖 Check out these 4 key concepts 🧵

Tweet media one

4

13

32

@weaviate_io

Weaviate • vector database

1 year

🤩 Are you attending #HaystackUS 2023 this week in Charlottesville, VA? 🗣️ Join @ecardenas300 from Team #Weaviate as she talks about Building Recommendation Systems with Vector Search on April 26. 🎟️ Don't miss this great talk! Get your tickets here:

Tweet media one

3

7

33

@weaviate_io

Weaviate • vector database

2 years

📣 welcome @bernhardsson to @SeMI_tech 's advisory board 🎆 As one of the early pioneers bringing ANN-based search end-to-end to users, we are honored to work with him on encapsulating his knowledge into @weaviate_io

Tweet media one

2

4

33

@weaviate_io

Weaviate • vector database

1 month

What if you could effortlessly access all your scattered information over different apps with ease? From important links in Slack to comments in notion to emails in Gmail, information is scattered across multiple places. This leads to time wasted on searching rather than…

2

9

33

@weaviate_io

Weaviate • vector database

6 months

Weaviate 1.22 is here! 💫 Discover the latest upgrades: • Nested JSON object storage • Async vector indexing • Additional gRPC support • v4 Python client beta Read the release blog post here: Explore the release now on WCS:

Tweet media one

1

10

33

@weaviate_io

Weaviate • vector database

11 months

Weaviate was always about scale. But lately, many of you required a different kind of scale: Running a single Weaviate set up for millions of tenants. Read @etiennedi 's preview of how millions of tenants will soon become a reality with Weaviate v1.20.

Tweet card media

Multi-Tenancy Vector Search with millions of tenants | Weaviate - Vector Database

Learn about the new native multi-tenancy feature

2

6

31

@weaviate_io

Weaviate • vector database

4 months

Multimodal search is as easy as 1, 2, 3 in Weaviate 🤩 1) Create a collection that can store audio, images, and videos. 2) Upload your data into the Weaviate collection you just defined. This creates a unified embedding space of all the modalities. 3) Now you can search! Follow…

Tweet media one

1

12

32

@weaviate_io

Weaviate • vector database

8 months

🚀 BIG NEWS: We're excited to partner with @streamlit for the LLM Hackathon, starting today! Build an innovative LLM-based Streamlit app to win amazing prizes! 📅 September 5-19, 2023 Dive in now:

Tweet media one

0

12

32

@weaviate_io

Weaviate • vector database

1 month

Large Language Models have significantly revolutionized the way we handle, store, and search through data. Their ability to quickly and affordably generate and modify information has improved user experiences through personalized content. We’ve gathered 5 terms/concepts to help…

Tweet media one

2

6

30

@weaviate_io

Weaviate • vector database

1 year

🚢 It's shipping time! 🔥 Weaviate Cloud Service has a newly redesigned management console! It offers managed Weaviate instances on the ☁️, and you can now create fully managed Weaviate clusters with minimal input! 🚀 Get started here: More in the 🧵 !

1

6

30

@weaviate_io

Weaviate • vector database

1 month

How can you reduce your Weaviate memory usage by 32x? Learn how binary quantization(BQ) works and how it can be used to enable brute force vector search, reduced memory usage, and faster indexing times. Read @ZainHasan6 and @AbdelRo68071159 's blog post, where they also share…

Tweet media one

1

8

30

@weaviate_io

Weaviate • vector database

2 years

🔥🔥🔥 📹 Another amazing interview on the Weaviate Podcast. This time @CShorten30 interviews @OpenAI 's @arvind_io ! Since OpenAI released their embeddings API, the integration is available as an OOTB module in the Weaviate vector search engine!

Tweet card media

Weaviate Podcast #7 • Arvind Neelakantan of OpenAI • Embeddings API...

Arvind Neelakantan, Research Lead at Open AI, talks with Connor Shorten about their newly released embeddings API, his work at Open AI, the integration into ...

www.youtube.com

3

10

30

@weaviate_io

Weaviate • vector database

3 months

The multi2vec-bind module, powered by @Meta 's ImageBind, enables Weaviate to generate vector data from various sources, including images, videos, audio, IMU (Inertial Measurement Unit) data, and single-channel depth images. What happens when we cross that with Next.js? Find…

0

6

29

@weaviate_io

Weaviate • vector database

5 months

Are you at NeurIPS in New Orleans? Look for @cshorten30 and @ecardenas300 for Weaviate stickers and an on-the-go short podcast! 😎

Tweet media one

4

5

28

@weaviate_io

Weaviate • vector database

8 months

We are beyond excited about the AutoGPT Arena Hacks starting today! 🚀 🤖 Create your AI agent powered by AutoGPT with Weaviate. 🔗 Join now and check out the **Hackathon Kick-Off Stream**: Read more 🧵↓ @Auto_GPT @lablabai

3

8

28

@weaviate_io

Weaviate • vector database

5 months

Weaviate 1.23, our last release of 2023, is live now! 🚀 Here’s what’s included: 1️⃣ AutoPQ for effortless vector indexing 2️⃣ Flat vector index and Binary Quantization, a new index for small collections 3️⃣ Open source LLM integration with generative-anyscale 4️⃣ Python client…

Tweet media one

0

8

26

@weaviate_io

Weaviate • vector database

1 year

🔥 Spotted Weaviate in @fireship_dev 's video on building an image search engine 🤩 Check it out and see how you can build your own using Weaviate! #ImageSearch #AI #MachineLearning

2

5

27

@weaviate_io

Weaviate • vector database

3 months

Reranking is a quick way to boost search relevance in your RAG application ☝️ We have a @cohere reranker module that supports the following models: • rerank-english-v2.0 • rerank-multilingual-v2.0 Test it out in this recipe demo:

Tweet media one

1

6

28

@weaviate_io

Weaviate • vector database

2 years

💡 New article by @laura_hamham : "Using Cross-Encoders as reranker in multistage vector search"

2

5

27

@weaviate_io

Weaviate • vector database

1 year

🔥 Vectors on disk (DiskANN) is coming to Weaviate! 📉 The graph (by @AbdelRo68071159 & @trengrj ) displays the memory drop after Weaviate writes the vectors from memory to disk 👀 Note how we use SSD instead of RAM! RAM dropped from 1.2 GB to 0.2 GB 🧵More below

Tweet media one

3

6

26

@weaviate_io

Weaviate • vector database

2 months

Weaviate 1.24 is now available! This new version is packed with amazing features and enhancements, including: 1️⃣ Named vectors 2️⃣ HNSW and binary quantization (BQ) 3️⃣ Simplified Docker configuration 4️⃣ Backend improvements 5️⃣ Python client update For more details, check out our…

8

8

26

@weaviate_io

Weaviate • vector database

1 year

👋 We'd like you to meet our new Developer Growth Team, with Philip Vollet ( @philipvollet ) and Edward Schmuhl ( @aestheticedwar1 ) 🥳 We can't wait to see their contributions. Welcome to Team Weaviate! 🎉 #Weaviate #DeveloperGrowth #TeamWeaviate

Tweet media one

1

8

26

@weaviate_io

Weaviate • vector database

1 year

🥳 This week, we published a demo on combining @LangChainAI and #Weaviate 📚 📹 Or watch @ecardenas300 's demo: 🧵 Check out the 4 CombineDocument techniques implemented in #LangChain that solve the limited input token lengths👇

Tweet card media

Weaviate + LangChain for LLM apps presented by Erika Cardenas

We are super excited to publish this overview of how LangChain and Weaviate work together from Erika Cardenas! Erika explains some of the key concepts behind...

www.youtube.com

4

8

27

@weaviate_io

Weaviate • vector database

9 months

What is a Vector Database? Welcome to Weaviate's most frequently asked question! This visual companion explains the core concept. Would you like to learn more? ✍️ Dive in "A Gentle Introduction to Vector Databases" by @helloiamleonie :

Tweet media one

1

8

26

@weaviate_io

Weaviate • vector database

13 days

Looking for a way to generate multimodal datasets for your demos, Proof of Concept (PoC), or project? Introducing Generator9000! Our latest open source tool allows you to create synthetic data tailored to your specific use case. Discover how you can seamlessly integrate…

0

4

26

@weaviate_io

Weaviate • vector database

2 months

Weaviate is an AI native open source vector database! This means you can deploy and use Weaviate however you’d like, but this comes with challenges. In Weaviate 1.24, we are releasing anonymized minimal telemetry to better understand how we can build the best version of Weaviate…

0

8

26

@weaviate_io

Weaviate • vector database

1 year

🪄 This week, we released the generative search module 💡 It allows #ChatGPT to use Weaviate search results to answer GraphQL queries based on content 🚀 Here is an example of a GraphQL query using the singleResult{ prompt } More in the 🧵

Tweet media one

1

9

26

@weaviate_io

Weaviate • vector database

3 years

Now also available in the Weaviate vector search engine. Awesome work @Nils_Reimers and co!

@Nils_Reimers

Nils Reimers

3 years

🚨Model Alert🚨 🏋️‍♂️ State-of-the-art sentence & paragraph embedding models 🍻State-of-the-art semantic search models 🔢State-of-the-art on MS MARCO for dense retrieval 📂1.2B training pairs corpus 👩‍🎓215M Q&A-training pairs 🌐Everything Available: 🧵

Tweet media one

Tweet media two

Tweet media three

21

85

429

1

8

26

@weaviate_io

Weaviate • vector database

6 months

Behind the scenes as we geared up for our Weaviate meetup in Bangkok today! Huge thanks to everyone who joined—it was a pleasure meeting you in person! 💚

2

3

26

@weaviate_io

Weaviate • vector database

19 days

Transforming data scattered across various sources into actionable insights, like searching or building chatbots, involves cleaning, formatting, setting up AI models, and other complex tasks. @unbody_io used Weaviate to build the foundation for AI-first app development, creating…

Tweet media one

0

7

24

@weaviate_io

Weaviate • vector database

9 months

Join us for the @DataCamp webinar with @_jphwang • Learn how vectors represent meaning • Discover the incredible capabilities of vector databases • Master Weaviate to perform semantic searches 🗓️ August 15, at 11 AM ET, 17 PM CET, 8 AM PT Register:

Tweet card media

Vector Databases for Data Science with Weaviate in Python

In this live training, JP shows you how to use Weaviate, a leading open source vector database, to build apps that can understand and manipulate them based on meaning.

www.datacamp.com

0

10

26

@weaviate_io

Weaviate • vector database

1 year

📉 Reducing memory footprint in vector databases comes with a choice between two trade offs: increased latency or reduced recall 🧵 Learn why and how to make your choice! 📚 Read about @ro48745762 's research in ANN algorithms and Product Quanitization

Tweet card media

HNSW+PQ - Exploring ANN algorithms Part 2.1 | Weaviate - Vector Database

Implementing HNSW + Product Quantization (PQ) vector compression in Weaviate.

2

8

26

@weaviate_io

Weaviate • vector database

7 months

It’s happening! We’re headed to Las Vegas Nov 27th - Dec 1st for @AWS_Partners re:Invent! Get a sneak peek:

Tweet media one

1

5

25

@weaviate_io

Weaviate • vector database

1 year

🤯 You can even use Weaviate's generative @OpenAI search module to summarize your search results! 💡 This code snippet shows you how to find the closest article in the database to the "bitmap index" and delivers the summary and the matching item 🚀 Check out the example below!

Tweet media one

1

5

25

@weaviate_io

Weaviate • vector database

2 years

🥳 this weekend, we've reached 1,000,000 downloads from 142 countries of the Weaviate vector search engine! More cool stats tomorrow…

Tweet media one

1

7

25

@weaviate_io

Weaviate • vector database

9 months

Dive deep into the world of Hybrid Search with Weaviate's latest Fusion Algorithms! This post shows you how you can harness the power of vector and keyword searches to perform a "best-of-both-worlds" search. Read the blog post:

Tweet media one

2

7

25

@weaviate_io

Weaviate • vector database

8 months

How to Reduce Memory Requirements with Product Quantization? Learn how to slash memory requirements by over 90% with PQ in our latest blog by @ZainHasan6 and @AbdelRo68071159 . Plus, master the rescoring trick to reduce the loss of recall. 👉

Tweet media one

2

10

24

@weaviate_io

Weaviate • vector database

2 years

📣 You can now use the Weaviate vector search engine in combination with @JinaAI_ 's DocArray out-of-the-box! Isn't that cool? Get started here:

Tweet media one

0

10

25

@weaviate_io

Weaviate • vector database

1 year

🚀 Join @Etienne ( @weaviate_io ) and @Nils_Reimers ( @CohereAI ) on Vector Search and Embeddings in Organizations 💻 Participate virtually tomorrow, March 30th, at 10 am PST, 1 pm EST, and 7 pm CET 😄 You can register for the event here: 🎉 See you there!

Tweet media one

0

10

24

@weaviate_io

Weaviate • vector database

11 months

What does a Vector DB & LLM Hackathon look like? Experience the excitement and innovation of this unforgettable Berlin Hackathon in collaboration with @mlopscommunity ! 🚀 🎬 Check out the full after-movie here:

1

7

25

@weaviate_io

Weaviate • vector database

1 year

🎉 We look forward to tonight’s @huggingface Open Source AI Meetup with 🤗 CEO in SF! 💡 Weaviate’s @dandv will be there to share his thoughts about open source vector DBs and answer all your questions 🤖 Join us, and let's talk tech! 👉

Tweet card media

Open Source AI meetup with 🤗 CEO | Partiful

Thanks for joining 🌟🔥🦙 In the coming weeks, we will share some of the open-source models, datasets, demos and research papers that were showcased & will come out of the event so make sure to...

1

8

24

@weaviate_io

Weaviate • vector database

3 months

v4 of the @weaviate_io #Python client has just been released! A lot has changed here, especially if you're coming from the v3 background. Learn how to create collections with the new syntax - where to import classes, add collection properties, vectorizers, RAG modules, index…

1

10

24

@weaviate_io

Weaviate • vector database

10 months

Want to combine multiple indexes into one? Discover the power of @llama_index 's composable graph! 🧩 You can create a list index from three different indexes, resulting in quick retrieval of relevant documents. Video: Notebook:

Tweet media one

0

8

24

@weaviate_io

Weaviate • vector database

2 months

Self-query retrieval in Healthsearch uses Large Language Models (LLMs) to interpret natural language queries and transform them into structured GraphQL queries. Query Translation LLMs translate natural language queries into structured queries, specifying the appropriate class…

0

4

24

@weaviate_io

Weaviate • vector database

1 month

#GoogleCloudNext is happening in just a few days! @bobvanluijt , Co-Founder and CEO of @weaviate_io , and @ebkauf , Senior Product Manager at @googlecloud , have a session on From RAG to autonomous apps with Weaviate and Gemini on Google Kubernetes Engine on April 11th at 10:15am…

Tweet media one

0

4

24

@weaviate_io

Weaviate • vector database

1 year

🎉 @OpenAI releases a New and Improved Embedding Model 📣 You can now use the new model (text-embedding-ada-002) in Weaviate! 👉 Docs: 🧵 For the improvements

3

8

24

@weaviate_io

Weaviate • vector database

3 months

Wait, are regular RAG pipelines a thing of the past? Why should you keep an eye on classic encoder models? This and more is part of our discussion with Aamir ( @aaxsh18 ) from 🍞 Listen to Aamir's take on the future of AI in this short interview! Also…

2

9

24

@weaviate_io

Weaviate • vector database

8 months

Weaviate is going to the @AIconference 2023! Join @etiennedi , Co-Founder & CTO of Weaviate, upstairs in the Robertson room tomorrow at 4:30 PM. Talk topic: "Solving multi-tenancy in vector search requires a paradigm shift." See you there!

Tweet media one

0

8

23

@weaviate_io

Weaviate • vector database

11 months

@etiennedi talks about the Weaviate Community at @berlinbuzzwords 👀 #bbuzz

3

11

23

@weaviate_io

Weaviate • vector database

24 days

Did you know, we simplified how you can start up a Weaviate instance with Docker in v1.24.1? Now you can run a basic Weaviate instance with the following command. The default docker image doesn't need any configuration! So, you don’t have to configure a docker-compose.yml file.…

Tweet media one

2

5

23

@weaviate_io

Weaviate • vector database

3 months

Product quantization compresses vectors to decrease memory usage up to 80%. PQ is a lossy algorithm, meaning that the compressed vectors have lost a lot of the information associated with them, so how does Weaviate still maintain good accuracy? First, a rough search is performed…

0

7

23

@weaviate_io

Weaviate • vector database

3 months

💫 Product Quantization (PQ) compresses vectors by breaking them down into smaller segments and then quantizing each segment independently. This process is similar to rounding a number to the nearest integer, but applied to vector spaces. 1) The original high-dimensional vector…

0

8

24

@weaviate_io

Weaviate • vector database

1 year

🤔 Curious about how to improve your search results? 📈 Ranking models are a great way to do exactly this ✍️ Read this blog post by @CShorten30 and @ecardenas300 to learn about the different ranking models (cross encoders, metadata rankers, score rankers)

Tweet card media

Ranking Models for Better Search | Weaviate - Vector Database

Learn about the different ranking models that are used for better search.

0

7

24

@weaviate_io

Weaviate • vector database

11 months

Weaviate has been recognized in @CBinsights ' AI 100 list for 2023! We're honored to be mentioned in the vector database tech category alongside other incredible innovators. Thanks to @CBinsights , our amazing team, and the Weaviate community. #AI100

Tweet card media

The top AI companies in 2023 | CB Insights

The AI 100 is CB Insights' annual list of the 100 most promising private AI companies in the world. This year’s winners are working on generative AI infrastructure, emotion analytics, general-purpose...

www.cbinsights.com

3

10

22

@weaviate_io

Weaviate • vector database

10 months

Want to visualize your embeddings? 🪄 This is now easier than ever with the new integration of Weaviate and Atlas from @nomic_ai Learn more about Nomic Atlas and what they are building in the 58th Weaviate Podcast with @andriy_mulyar and @CShorten30 !

Tweet card media

Andriy Mulyar on Nomic AI, Atlas, and GPT4All - Weaviate Podcast #58!

Hey everyone! Thank you so much for watching the 58th episode of the Weaviate Podcast! I am SUPER excited to welcome Andriy Muylar! Andriy is the Co-Founder ...

www.youtube.com

0

8

23

@weaviate_io

Weaviate • vector database

7 months

Dive into the world of knowledge assistants with Weaviate! This hackathon is centered around @cohere 's Coral. Explore how it's built using Chat and RAG, create innovative apps, and win awesome prizes. With @cohere , @LangChainAI , and @lablabai Apply:

1

7

23

@weaviate_io

Weaviate • vector database

1 year

🧠 Want to learn more about GPT Index? 🎧 Listen to this podcast with @gpt_index 's Creator @jerryjliu0 , and @bobvanluijt and @CShorten30 from Team Weaviate 🚀 We discuss how LLMs are powering new ways to Search in GPT Index 😄 Enjoy listening!

Tweet card media

GPT Index and Weaviate with Jerry Liu and Bob van Luijt - Weaviate...

Hey everyone! Thank you so much for watching the 37th episode of the Weaviate podcast! This episode discusses some of the ideas behind GPT Index. GPT Index p...

www.youtube.com

1

9

23

@weaviate_io

Weaviate • vector database

8 months

🎤 Join the session with @philipvollet , Head of Developer Growth at Weaviate, as he presents "Intro to Vector Databases." Enroll now and join the talk: See you there, and let's make this Hackathon an incredible experience!

Tweet media one

1

7

23