Try Milvus 2.4 Features in Zilliz Cloud, Learn about vector embeddings, Check out an AI podcast, and more!
In this issue:
- Milvus 2.4 New Features Available in Beta on Zilliz Cloud
- An Introduction to Vector Embeddings
- AI Podcast Recommendation
- ICYMI: Unstructured Data Meetup Recaps
- Upcoming Events
📣 Milvus 2.4 New Features Available in Beta on Zilliz Cloud
In the latest release of Zilliz Cloud, you can try Milvus 2.4 features and more:
🔎 Multi-vector and hybrid search - enhanced!
⏩ Inverted index and fuzzy matching - faster queries!
🔢 Grouping search - document-level recall!
🌊 Float16 and BFloat16 data types - improved search efficiency!
Milvus 2.4 is available in beta on the free, Serverless, and dedicated clusters. If you’re using dedicated clusters, click upgrade to explore exciting new features.
A Pipelines update:
🔍 Expanded Search Functionality: Pipelines are now even more powerful, supporting searches across text, documents, and images.
🔢 An Introduction to Vector Embeddings
There’s neural networks, matrix factorization, GloVE, BERT, ColBERT, SPLADE, and BGE-M3. So many embedding models out there, but what are the differences? In this blog post, we will understand the concept of vector embeddings and explore its applications, best practices, and tools for working with embeddings. Follow the step-by-step approach at the end to store, index, and search vector embeddings with Milvus.
🎙️AI, Vector Databases, and Data Sovereignty with Stephen and Vinay
Listen to Stephen Batifol on the Severalnines show with Vinay Joosery discussing how ChatGPT and other LLMs have amplified interest in vector databases. The following are some hot 🔥 topics discussed in the show:
⚡ A second act: Vector databases have never been more relevant
⚡LLMs as Magwai: Be careful how you care for them
⚡Keep your data close, and your AI closer: How to ensure data privacy
⚡The devil you know: When to go augmented or specialized
🎥 Unstructured Data Meetup Recaps
If there is no meetup program in your city or you missed an interesting one, check out the recap blogs and replays on YouTube!
🥪 How Delivery Hero Implemented the Safety System for AI-Generated Images
Learn how two data scientists at Delivery Hero use AI models to generate high-quality food images to improve user experience and conversion rate. Their approach consists of two stages: food image generation and building a safety system.
▶️ Slides
🔎 Vector Search and RAG - Balancing Accuracy and Context
Building RAG requires careful choices of embedding models, indexes, and semantic search approaches. By addressing the issue of AI hallucinations and leveraging dynamic retrieval of up-to-date information, RAG offers a powerful tool for creating more reliable and context-aware AI systems.
▶️ Get started (notebook)
💻 Using Vector Search to Better Understand Computer Vision Data
In this talk, you will learn how to combine vector search engines with the Voxel51 FiftyOne open source computer vision library for unstructured data curation and visualization, so you can interactively explore and find hidden structures in your data. From standard applications like similarity search and reverse image search, to multimodal applications such as semantic search and concept interpolation, you’ll see your data like never before.
▶️ Get started (tutorial)
👗 Elevating User Experience with Image-Based Fashion Recommendations
Learn about an innovative AI system that is revolutionizing fashion retail with personalized outfit recommendations. By combining visual embeddings and vector databases, this approach uses a PyTorch autoencoder and YOLOv5 for fast, accurate suggestions. The web demo app showcases these capabilities, providing real-time, tailored fashion advice.
🔢 Training Text Embeddings with Jina AI
Embeddings have become a crucial component in contemporary vector search and Retrieval Augmented Generation (RAG) systems. In this talk, Bo Wang from Jina AI provides a comprehensive overview of training a versatile embedding model, strategies for encoding longer information within such models, along their benefits and limitations. Additionally, he explains various forms of deep learning-powered retrievers.
▶️ Slides
Upcoming Events
July 4: Albania Tirana Tech Meetup (in-person)
Join Stephen Batifol at the next Tirana Tech Community Meetup in Albania! Don’t miss out on this opportunity to expand your knowledge and connect with fellow tech enthusiasts. Whether you’re a developer, entrepreneur, or simply curious about the latest in tech trends, this event is for you!
July 8 - Aug 3: Backdrop Build Hackathon (virtual)
We're excited to be a part of Backdrop Build to support builders on Milvus and Zilliz! Join from anywhere in the world for a 4-week online program to build and launch one of your cool ideas with support 🤝
July 9: Brazil Unstructured Data Meetup (in-person)
Join us for our first meetup in Brazil featuring Frank Liu speaking about “How Vector Databases are Revolutionizing Unstructured Data Search in AI Applications.”
Watch the livestream 3:00 - 4:00 PM (Local time - Recife) / 11 AM - 12 PM (Pacific Time)
July 11: Building an Agentic RAG locally with Milvus, Ollama and LangGraph (virtual)
RAG systems are talked about in detail, but usually stick to the basics. In this talk, Stephen Batifol will show you how to build an Agentic RAG System using Langchain and Milvus.
July 16: San Francisco Unstructured Data Meetup (in-person)
Join us in San Francisco with APARAVI and Encord for a meetup on July 16! Join us for the following talks:
▶️ Towards Unstructured Data with Zilliz CEO, Charles Xie
▶️ Garbage In, Garbage Out: Why poor data curation is killing your AI models (and how to fix it) with Alexandre Bonnet from Encord
▶️ It's your unstructured data: How to get your GenAI app to production (and speed up your delivery) with Joe Maionchi and Hendrik Krack from Aparavi
Can’t make it in person? Join us virtually on Twitch: https://www.twitch.tv/vectordatabase
July 18: RAG Evaluation with Ragas (virtual)
This talk will demo ragas , an open-source automation tool for RAG evaluations. Christy Bergman will talk about and demo evaluating a RAG pipeline using Milvus and RAG metrics like context F1-score and answer correctness.
July 25: Hands-On Demo: Building and Scaling Vector Search Apps with Zilliz Cloud (virtual)
This session will teach you how to implement efficient, scalable search solutions for your GenAI apps. Frank Liu will walk you through building and scaling vector search applications with live examples. By the end, you'll have a clear understanding of how to leverage Zilliz Cloud in your projects and hands-on experience with its capabilities.
July 25: New York Unstructured Data Meetup (in-person)
Join us in New York at the Cloudera office for a meetup on July 25! Hear 3 exciting talks:
▶️ Unstructured Data Processing From Cloud to Edge, Tim Spann 🥑 , Zilliz
▶️ RAG Pipelines with Apache NiFi, Chris Joynt , Cloudera
▶️ Metadata Lakes for Next-Gen AI/ML, Lisa N. Cao , Datastrato
VP of Marketing @ Zilliz (creators of Milvus, world's most popular open-source #VectorDatabase). AI Alliance member. Loves stories from developers on building RAG apps, product recommenders #AI #LLM
2wThanks for sharing