Timescale

PostgreSQL and pgvector: Now Faster than Pinecone, 75% cheaper, 100% open-source. Introducing pgvectorscale, an open-source PostgreSQL extension that builds on pgvector, enabling greater performance and scalability. Here’s how pgvectorscale helps pgvector outperform specialized vector database like Pinecone: 1️⃣ StreamingDiskANN:  A new vector search index that overcomes limitations of in-memory indexes like HNSW the index on disk, making it more cost-efficient to run and scale as vector workloads grow. Inspired by the DiskANN paper from Microsoft. 2️⃣ Statistical Binary Quantization (SBQ): Developed by researchers at Timescale, this technique improves on standard binary quantization techniques by improving accuracy when using quantization to reduce the space needed for vector storage 3️⃣ Written in Rust, giving the PostgreSQL community to contribute to vector support. 📈The result? On our benchmark of 50 million Cohere embeddings (768 dimensions each), PostgreSQL with pgvector and pgvectorscale achieves 28x lower p95 latency and 16x higher query throughput compared to Pinecone for approximate nearest neighbor queries at 99 % recall, all at 75 % less cost when self-hosted on AWS EC2. We also tested it against Pinecone’s p2 high performance index, see the blog post at the end of this post for full results (spoiler: It’s just as impressive). Pgvectorscale is open-source under the PostgreSQL license and free for you to use on any PostgreSQL database for your AI projects. To get started, see the pgvectorscale github repo: https://lnkd.in/ghXj2e-U Or try it on Timescale Cloud on any new database service. Eager to learn more about pgvectorscale and how it works? Head over to our blog post with all the details: https://lnkd.in/gcMcxrVb

Shocker: Postgres is all you need for RAG Great tutorial from Dave Ebbelaar on how to build high performance RAG using pgvector and pgvectorscale (Link in comments) #rag #pgvector #vectorsearch

Shocker: Postgres is all you need for RAG Great tutorial from Dave Ebbelaar on how to build high performance RAG using pgvector and pgvectorscale (Link in comments) #rag #pgvector #vectorsearch

Learn how Pondhouse Data OG built a Content Recommendation System for SEO purposes fully on top of PostgreSQL, using extensions pgai and pgvectorscale. 🤯👇 Pondhouse Data OG specializes in customized AI applications, from self-hosted chat tools to tailored AI model training, including content recommendation systems. As a small team, they prioritize minimizing complexity while leveraging powerful technologies like large language models (LLMs) and vector search, especially for retrieval-augmented generation (RAG). Andreas Nigg, CEO & Co-founder of Pondhouse Data OG, shares his insights into building this system designed for SEO-driven internal link building, utilizing pgai and pgvectorscale. These tools empowered Pondhouse Data OG to meet the challenges of multi-tenancy and data privacy head-on, ensuring seamless AI tool resale for their clients. By integrating these extensions, Pondhouse was able to reduce infrastructure complexity while still delivering robust, secure solutions. This hands-on guide walks you through the step-by-step process, demonstrating how pgai and pgvectorscale can simplify the development of AI applications within PostgreSQL. ✨ Check out why these tools are among the best solutions for building scalable, secure AI systems with minimal infrastructure setup and how Pondhouse Data OG successfully harnessed their power. 💪 🔗 Link below to learn more.

🧙♂️ Magic of AI meets Database Magic! 🪄 I was showing my kids Claude (an AI assistant) and we asked it to create a Harry Potter spell to make the world adopt Timescale. The result was hilariously creative and surprisingly on-brand: Spell: "Timescalus Adoptus Mundus!" Effect: Creates dazzling time-series data streams circling the globe, transforming into Timescale logos as they touch continents. It instills an urge in DBAs and developers worldwide to explore Timescale. My favorite part? The cautionary note about side effects, including "an irresistible urge to optimize database performance" and "unconsciously doodling the Timescale logo in meetings." 😂 Who knew AI could blend wizardry with database marketing so seamlessly? Maybe we should add this spell to our official documentation! 🤔 #AI #Timescale #HarryPotter #TechHumor #DatabaseMagic

𝙋𝙧𝙤𝙣𝙪𝙣𝙘𝙞𝙖𝙩𝙞𝙤𝙣: 𝘵𝘪𝘮𝘦-𝘚𝘒𝘈𝘓-𝘶𝘴 𝘢-𝘋𝘖𝘗-𝘵𝘶𝘴 𝘔𝘜𝘕-𝘥𝘶𝘴 𝙒𝙖𝙣𝙙 𝙢𝙤𝙫𝙚𝙢𝙚𝙣𝙩: 𝘋𝘳𝘢𝘸 𝘢 '𝘛' 𝘧𝘰𝘳 𝘛𝘪𝘮𝘦𝘴𝘤𝘢𝘭𝘦, 𝘵𝘩𝘦𝘯 𝘵𝘳𝘢𝘤𝘦 𝘢 𝘤𝘶𝘳𝘷𝘦𝘥 𝘢𝘳𝘳𝘰𝘸 𝘵𝘰 𝘳𝘦𝘱𝘳𝘦𝘴𝘦𝘯𝘵 𝘢𝘥𝘰𝘱𝘵𝘪𝘰𝘯, 𝘢𝘯𝘥 𝘧𝘪𝘯𝘪𝘴𝘩 𝘸𝘪𝘵𝘩 𝘢 𝘤𝘪𝘳𝘤𝘶𝘭𝘢𝘳 𝘮𝘰𝘵𝘪𝘰𝘯 𝘳𝘦𝘱𝘳𝘦𝘴𝘦𝘯𝘵𝘪𝘯𝘨 𝘵𝘩𝘦 𝘸𝘰𝘳𝘭𝘥. 😂 AI is the new playground

🚀 In this Issue: PostgreSQL AI = Magic! 🎩 🪄 We’re diving deep into how PostgreSQL is powering the future of AI-driven applications. Get started with our guest blog from the Pondhouse Data OG team, where they share how they’re using AI to build a smart content recommendation system. Curious about AI for visual projects? Discover how you can create an AI-powered image gallery using OpenAI CLIP, Claude Sonnet 3.5, and pgvector. It’s a can’t-miss guide for anyone looking to combine PostgreSQL with cutting-edge AI tools. And don’t forget—another Launch Week is coming soon, where we’ll introduce powerful new features to boost reliability and simplify developer workflows! 💪 Read more:

Don't sleep on pgai 🔥🔥

If I told you Postgres could directly connect to AI models, eliminating messy infrastructure and turning 150-line Python scripts into single SQL queries, you'd probably think I'm crazy. But that's exactly what pgai does. Andreas Nigg, AI engineer and pgai user, explains more: #ai #llm #postgresql #rag #vectorsearch

If I told you Postgres could directly connect to AI models, eliminating messy infrastructure and turning 150-line Python scripts into single SQL queries, you'd probably think I'm crazy. But that's exactly what pgai does. Andreas Nigg, AI engineer and pgai user, explains more: #ai #llm #postgresql #rag #vectorsearch