Grafana Labs launched Grafana 13 at GrafanaCON 2026, anchored by a rebuilt next-generation architecture for Grafana Loki, its log-aggregation system, along with a deliberately simplified path to getting OpenTelemetry instrumentation running on Linux and Kubernetes hosts. The Loki rework matters because log storage and query performance at scale has historically been Loki's biggest pain point compared to more heavyweight systems like Elasticsearch; a next-generation architecture aimed squarely at that bottleneck signals Grafana Labs is trying to close the gap that has pushed some teams toward pricier, more resource-intensive alternatives purely for query performance reasons. On the OpenTelemetry side, the release leans into a broader industry shift: OpenTelemetry has become the de facto standard for cloud-native instrumentation, but actually wiring up a collector correctly across a fleet of Linux hosts or a Kubernetes cluster has remained fiddly enough that plenty of teams still put it off. Grafana 13 also brings dynamic dashboards to general availability, letting a single dashboard adapt its layout and content based on variables and context rather than requiring a hand-built dashboard per team or per service, plus dashboard layout templates aligned to standard methodologies like DORA metrics and the USE/RED method, giving teams a starting point instead of a blank canvas. For platform and SRE teams evaluating their observability stack, the practical takeaway is that Grafana is explicitly targeting the two biggest adoption-friction points in open observability, expensive log query performance and painful OpenTelemetry setup, rather than only adding new visualization features, which is the more durable kind of investment for teams trying to reduce the operational burden of running observability infrastructure themselves.