Scaling retrieval shouldn’t break your budget.
OpenSearch supports Faiss byte vectors, cutting storage and memory costs while keeping retrieval performance strong.
Save memory. Reduce costs. Scale faster. https://opensearch.org/blog/faiss-byte-vector/
#VectorSearch #OpenSearch #AI #ML #SearchTechnology #SmartSearch #OpenSource
One week away → Chicago's Search User Group!
Join for an evening of OpenSearch insights and vector search case study deep dive.
Tuesday, August 19, 2025 | 6:00 PM to 8:00 PM CDT
Chicago
Register now 
https://www.meetup.com/opensearch-project-chicago/events/310125523/
Structured data powers your applications.
Now it can power your vector search too.
Learn how to generate embeddings from nested objects — and extend vector retrieval beyond plain text.
Smarter hybrid search starts here. https://docs.opensearch.org/docs/latest/tutorials/vector-search/vector-operations/generate-embeddings/
#VectorSearch #OpenSearch #SmartSearch #OpenSource
Chicago's Search User Group!
Join for an evening of OpenSearch insights and vector search case study deep dive.
Tuesday, August 19, 2025 | 6:00 PM to 8:00 PM CDT Chicago
To learn more and to register https://www.meetup.com/opensearch-project-chicago/events/310125523/
LLMs don’t know your PDF.
They don’t know your company wiki either. Or your research papers.
What they can do with RAG is look through your documents in the background and answer using what they find.
But how does that actually work? Here’s the basic idea behind RAG:
Chunking: The document is split into small, overlapping parts so the LLM can handle them. This keeps structure and context. Embeddings & Search: Each part is turned into a vector (a numerical representation of meaning). Your question is also turned into a vector, and the system compares them to find the best matches. Retriever + LLM: The top matches are sent to the LLM, which uses them to generate an answer based on that context.
@krisfreedain and Nils Bandener had a great conversation at the #WeAreDevelopers conference in Berlin, diving into the OpenSearch community, what it takes to become a maintainer, and the journey to getting there.
Thanks to the OpenSearch community for keeping the momentum going and pushing boundaries.
OpenSearch 3.1 delivers massive performance gains for hybrid queries with up to 3.5x higher throughput and 80% lower latency! Our new score collection system optimizes parallel execution for faster, more efficient searches. Check out the technical details here 
http://bit.ly/46CvLrO
At Open Source Summit North America, @theCUBE host Paul Nashawaty caught up with Mukul Karnik from OpenSearch to discuss key highlights from OpenSearch 3.0 and new features in 3.1. Check out their chat covering all things hybrid search, MCP, vector search, observability use cases and agentic AI.
Watch the full interview: https://www.thecube.net/events/linux-foundation/open-source-summit-na-2025/content/Videos/01a2abef-d221-486b-9eea-4cfdca4e97a8
OpenSearch made a strong showing at #WeAreDevelopers in Berlin.
Jochen Kressin, Director at @eliatra shared how they're using OpenSearch as a vector database, and how 3.0 features like GPU support are driving real impact.
Huge thanks to the OpenSearch community. The momentum is real and the future is bright.
In Berlin at #WWC25 ? Catch our 
newest Pragprog author - Ben Greenberg @hummusonrails
Learn how GenAI and Vector Search can help users find what they are looking for - even when they don't know!
Fri 2:20 pm - Stage 7
Ben's Book - out in Beta this week!
https://pragprog.com/titles/bgvector?utm_source=m
#AI #MachineLearning #VectorSearch
WeAreDevs
Celebrating Vector Search with JavaScript: Build Intelligent Search Systems with AI by Ben Greenberg now in beta!
Read more: https://medium.com/pragmatic-programmers/57eed26bc43d
OpenSearch 3.1 is here with powerful upgrades to boost search, speed up generative AI, and improve observability.
Highlights include:
GPU-accelerated vector search New Search Relevance Workbench Smarter agent and model management Simplified semantic search Better ML monitoring with OpenTelemetry
Explore what’s new and get started: https://opensearch.org/blog/get-started-with-opensearch-3-1/
"We’re really pushing the boundaries of search.” — Pallavi Priyadarshini, OpenSearch Project
In a packed keynote at #OpenSearchCon India, leaders from AWS and Freshworks shared how OpenSearch is powering real-world AI use cases—from semantic and hybrid search to federated agents and chat-based assistants.
To learn more, read the blog here: https://opensearch.org/blog/vector-powered-ai-ready-how-opensearch-and-freshworks-are-shaping-the-future-of-search/
Headed to #OpenSourceSummit North America in Denver? Don’t miss these OpenSearch sessions:
Keynote: Lucene 10 + OpenSearch 3.0 powering the next wave of AI search
Lightning Talk: From Fork to Foundation – the OpenSearch journey to @linuxfoundation
Full schedule: https://events.linuxfoundation.org/open-source-summit-north-america/program/schedule/
Vector search, now more efficient. 
OpenSearch 3.0’s derived source cuts storage up to 3x and lowers search latency by 90%.
No code changes. Just better performance.
Try it now https://opensearch.org/blog/do-more-with-less-save-up-to-3x-on-storage-with-derived-vector-source/
Smarter pipelines, better performance.
OpenSearch now skips redundant inference calls—saving compute, lowering costs, and speeding up updates.
Learn how it works: https://opensearch.org/blog/optimizing-inference-processors-for-cost-efficiency-and-performance/
Summoning all #dataengineers, #AI enthusiasts, and #RAG masterminds - join us at #VSCON25 June 6th to expand your knowledge and network with global developers from #Google, #Oracle, and #Microsoft!
Sign up at: http://vsearchcon.com
Keyword matching vs. semantic understanding: The battle between traditional & vector search. Discover why context wins for AI apps: https://opensearch.org/platform/vector-engine/ #VectorSearch #AI #OpenSource
ebook: 