OpenSearch version 3.8 is now available on ClusterNest

OpenSearch 3.8 is live on ClusterNest, with faster vector search, a broader set of AI agent integrations, and new tools for exploring logs without leaving Dashboards. Vector ingestion is up to 4.16x faster and radial search throughput is up to 2.1x higher, thanks to base64 vector encoding that also cuts network payload size significantly. On the AI side, Model Context Protocol support now extends to all four agent types, with a new Tool Discovery API and per-connector tool customization for better LLM tool selection. Streaming ML predictions can now run over gRPC for lower latency, and LLM-as-a-Judge evaluation adds support for several more providers, including Anthropic Claude and Amazon Bedrock. For log analysis, Dashboards gains a visual PPL query builder, experimental SQL support in Discover, and a handful of new PPL commands for reshaping and comparing data. See the OpenSearch 3.8 announcement for the full rundown.

Provisioning and managing 3.8 clusters is supported today in both the Console and the Terraform provider.

Try it out from the ClusterNest Console, or update your Terraform configuration to move existing clusters onto 3.8.