Blogs
Developer insights and practical how-to articles from our experts to inspire and empower your search experience.
Elasticsearch DiskBBQ: 40% faster vector scoring with native SIMD Blocks
A deep dive into how DiskBBQ's block layout, doc ID compression modes and native SIMD kernels combine to deliver 40% improved vector scoring throughput for DiskBBQ in 9.4.
Build a RAG agent with Elasticsearch and GitHub Copilot SDK
Wire Elasticsearch into the GitHub Copilot SDK as a RAG tool in five lines of C#, grounding your agent in your own logs, docs and data instead of model training data.
Multilingual image search with Jina CLIP v2 and Elasticsearch
Build a multilingual image search system using Jina CLIP v2 and Elasticsearch. Query your image collection in 89 languages with no translation pipeline, and use Matryoshka Representations to cut index size by 75%
One API call per operation: how Elastic Cloud Hosted makes fleet-scale deployment management practical
Elastic Cloud Hosted adds five targeted APIs for upgrade, tier scaling, user settings, tags and snapshot repository linking, each replacing a multi-step deployment plan edit with a single focused call.
How we doubled vector search throughput on Elasticsearch Serverless
How we brought Elasticsearch's native SIMD scoring engine to serverless, and why serverless is where vector search innovation happens next.
How Elasticsearch cuts time-series storage by 34% with synthetic _id and bloom filters
Learn how synthetic _id uses bloom filters to cut time-series storage by 34% while maintaining full API compatibility.
Small model, big benchmarks: how Jina-VLM beat the competition at 2.4B and what ICLR told us is coming next
Jina-VLM is a 2.4B open multilingual VLM leading VQA benchmarks across 29 languages. Plus: five days of ICLR 2026 takeaways on RLVR, sparse embeddings and retrieval.
Cutting Elasticsearch DiskBBQ query quantization time by 5x
See how asymmetric quantization cuts DiskBBQ query quantization overhead from about 20% to 4% with little recall impact.
Cutting agent costs with pre-computed context
Pre-computing context as Knowledge Indicators reduces LLM agent token costs by up to 75% and improves answer accuracy from 60% to 92%. This post covers the extraction, retrieval and feedback loop that make it work, tested against the BrowseComp-Plus benchmark.
Kibana dashboards as code: GitOps, drift detection and Terraform for Kibana dashboards in Elastic 9.4
Elastic 9.4 ships a typed Dashboards API and a native Terraform resource that bring drift detection, PR-reviewable diffs and git-based rollback to Kibana dashboards for the first time.
