Prometheus native histograms in Grafana Cloud: Get more precision from your Grafana visualizations

In May, we announced the public preview of Prometheus native histograms in Grafana Cloud, unlocking greater precision, ease of use, and compatibility for analyzing latency, duration, and other distributions. Since then, we’ve seen incredible adoption across industries—from financial services companies to e-commerce platforms. 

Last week, during PromCon EU 2025, the Prometheus developers announced that native histograms are now stable, after three years of intense testing and improvements. And today, we’re excited to share that Prometheus native histograms are also now generally available in Grafana Cloud!

Why customers chose native histograms

Compared to the classic Prometheus histograms, Prometheus native histograms—native histograms, for short—offer higher resolution and precision. They’re also easier to instrument and you can use them to combine and manipulate histograms in queries and in Grafana.

Plus, native histograms provide compatibility with other formats, such as OpenTelemetry exponential histograms and Datadog distributions.

During the public preview, our customers put native histograms in production-like scenarios. They found that traditional histograms, with their predefined bucket boundaries, often made it difficult to achieve the right level of precision. Native histograms solved this by dynamically adjusting buckets and efficiently capturing distributions.

Here’s what our customers told us:

• A financial services customer utilized native histograms to dynamically configure histogram buckets, thereby reducing the operational and monetary costs associated with histograms.
• An e-commerce platform onboarded dozens of internal teams without the need to configure the buckets manually each time. This has dramatically reduced the toil of manual onboarding and onboarded teams much faster. The customer also reduces the number of buckets by around 30% through a more efficient use of histograms. 

Note: Because native histograms are stable in Prometheus, that means they’re also stable in other downstream projects like Grafana Mimir.

What is new in GA

The biggest change since we first announced native histogram support is the direct integration with Grafana Metrics Drilldown, our popular queryless experience for browsing Prometheus-compatible metrics.

With instant access to the higher-resolution, more precise histogram data, the app automatically identifies native histogram metric types and generates the corresponding queries and visualizations, allowing you to view your data immediately without needing to write PromQL. 

This streamlined workflow enables you to visualize native histogram distributions immediately, while also comparing them side by side with classic histograms to understand the differences in resolution and accuracy. 

When you’re ready to dig deeper, simply select any native histogram metric and use the “Open in Explore” function to see the underlying PromQL functions, giving you a clear path from queryless exploration to advanced query customization.

Pricing

During the public preview, we found that our current pricing worked well for customers, and we intend to maintain this pricing structure going into GA.

If you haven’t used this feature yet, be aware that native histograms are billed differently from regular time series. A native histogram active series is equal to one active bucket multiplied by 0.25. We made this pricing change to pass on the resource savings to our customers. As a result, customers can leverage the high resolution from native histograms without worrying about cost. 

What’s next

We’re actively working on native histogram custom buckets (NHCB), which is an additional kind of native histogram for Prometheus that let users keep their existing histograms (no change to the instrumentation, only to the data collection) and store them in native histograms. 

The benefits include lower costs and increased reliability. We should note that NHCB depends on Prometheus Remote-Write 2.0, which is still experimental. NHCB also requires updates to queries the same way as other native histograms. We’ll share more documentation for using NHCB in Grafana Cloud in the future. 

We also plan to integrate native histograms directly with Grafana Cloud products like Application Observability and Frontend Observability, so that users will get more out-of-the-box experience across our cloud offerings.

And to get started with native histograms today, please refer to our documentation to learn more.

Grafana Cloud is the easiest way to get started with metrics, logs, traces, dashboards, and more. We have a generous forever-free tier and plans for every use case. Sign up for free now!