Skip to content
Monitoring

Prometheus Releases

Track Prometheus releases, TSDB evolution, remote write improvements, native histograms, and upgrade guidance. Thanos and Mimir comparison context for long-term storage decisions.

Official Site Documentation

—

Total Versions

—

Supported

—

Latest

Version Timeline

All tracked releases with lifecycle status and EOL dates.

Loading version data…

Lifecycle Timeline

Visual overview of active support and maintenance windows.

2.45 LTS
2.48
2.50
2.53 LTS
2.55
3.0
3.1
2023 2024 2025 2026 2027 2028
Active
Maint
Active
Maint
Active
Maint
Active
Maint
Active
Maint
Active
Maint
Active
Maint
Active / LTS
Maintenance
Today

Upgrade Paths

Migration guidance between major versions — breaking changes, effort estimates, and tips.

2.45 LTS 2.53 LTS Low Difficulty
Est. < 30 minutes

Breaking Changes

  • Native histograms now stable (opt-in)
  • Remote write v2 protocol available
  • Some deprecated CLI flags removed
  • Updated default scrape timeout behavior
  • TSDB format improvements (backward compatible)

Migration Notes

LTS-to-LTS upgrade is smooth. Replace binary, restart. TSDB format is forward-compatible. If you use --enable-feature flags, review which features graduated to stable. No PromQL breaking changes between these versions.

2.x (any) 3.0 Medium Difficulty
Est. 1-3 hours including dashboard verification

Breaking Changes

  • Major version: deprecated features removed
  • Native histograms enabled by default
  • Remote write v2 as default protocol
  • UTF-8 metric names support
  • OpenTelemetry-native ingestion improvements
  • Some experimental features graduated or removed
  • CLI flag cleanup

Migration Notes

The 2.x to 3.0 jump removes accumulated deprecations. The biggest impact is native histograms becoming the default and remote write v2. If your downstream consumers (Grafana, alerting rules) use classic histograms, verify compatibility. UTF-8 metric name support may affect regex-based relabeling rules. Test with your actual dashboards before switching.

Version Risk Assessment

Evaluate risk factors before choosing a version for production.

Version EOL Risk CVE Risk Ecosystem Cloud Support Overall Recommended Action
Prometheus 2.45 LTS High Medium EOL Full High LTS ended Dec 2024 — upgrade to 2.53 LTS or 3.0
Prometheus 2.48-2.52 High Medium Unsupported Full High No patches — upgrade to 2.53 LTS or 3.0
Prometheus 2.53 LTS Low Low Active Full Low Current LTS — recommended for stability
Prometheus 3.0 Low Low Active Full Low Current — recommended for new deployments
Prometheus 3.1 None Low Active Full Low Latest — cutting edge

Prometheus only patches the latest release (plus LTS). Running old versions means no security fixes for the monitoring system itself. Assessed March 2026.

Prometheus Version Feature Comparison

Side-by-side feature differences across major versions.

Feature 2.45 LTS 2.50 2.53 LTS 3.0
Native histograms Experimental Stable Stable Default
Remote write protocol v1 v1 + v2 opt-in v1 + v2 v2 default
UTF-8 metric names No Experimental Experimental Stable
OTLP ingestion Experimental Improved Stable Enhanced
TSDB format v2 v2 v2 v3
Scrape protocols OM 0.0.4 OM 1.0.0 OM 1.0.0 OM 1.0.0 + OTLP
Recording rules Stable Stable Stable Enhanced
Service discovery 30+ types 35+ types 35+ types 40+ types
Agent mode Stable Stable Stable Default option

Embed Badges

Add live Prometheus status badges to your README, docs, or dashboard.

Health Status

Overall support health

Prometheus Health Status
![Prometheus Health Status](https://img.releaserun.com/badge/health/prometheus.svg)

EOL Countdown

Next end-of-life date

Prometheus EOL Countdown
![Prometheus EOL Countdown](https://img.releaserun.com/badge/eol/prometheus.svg)

Latest Version

Current stable release

Prometheus Latest Version
![Prometheus Latest Version](https://img.releaserun.com/badge/v/prometheus.svg)

CVE Status

Known vulnerabilities

Prometheus CVE Status
![Prometheus CVE Status](https://img.releaserun.com/badge/cve/prometheus.svg)

Frequently Asked Questions

Common questions about Prometheus releases and lifecycle.

How does Prometheus versioning and support work?
Prometheus follows semantic versioning with minor releases roughly every 6 weeks. Only the latest minor version receives patches. There is no formal LTS. The project is conservative about breaking changes: PromQL, config format, and the remote write API are treated as stable. TSDB format changes are backward compatible (Prometheus can read older data formats).
What are native histograms?
Native histograms (stable since Prometheus 2.50+) replace classic histograms with a more efficient format that automatically adjusts bucket boundaries. They reduce storage by 10-100x for histogram data and eliminate the need to pre-configure bucket boundaries. This is the biggest observability improvement in Prometheus in years. Enable with --enable-feature=native-histograms.
Should I use Thanos, Mimir, or Victoria Metrics for long-term storage?
Thanos is the CNCF-graduated option: it adds a sidecar to Prometheus and stores data in object storage (S3, GCS). Good if you want to keep Prometheus mostly unchanged. Mimir (Grafana Labs) is a horizontally-scalable Prometheus-compatible TSDB that replaces Prometheus entirely for ingestion and storage. Better for very high cardinality. VictoriaMetrics is the performance-focused alternative with excellent compression. Choose Thanos for minimal change, Mimir for scale, VM for performance.
How does Prometheus remote write work?
Remote write sends scraped metrics from Prometheus to an external TSDB in real-time. Remote write v2 (Prometheus 2.54+) uses protobuf with metadata and native histogram support, reducing bandwidth by ~30-50% over v1. Most long-term storage solutions (Thanos Receive, Mimir, VictoriaMetrics, Cortex) accept remote write.
How do I upgrade Prometheus safely?
Prometheus upgrades are generally safe: stop Prometheus, replace the binary, start. TSDB format is backward compatible. The main risks are: PromQL behavior changes in edge cases (rare), config file key additions/deprecations, and scrape configuration changes. Always read the CHANGELOG. For zero-downtime, run two Prometheus instances behind a load balancer or use Thanos/Mimir.
What is the Prometheus operator?
The Prometheus Operator (kube-prometheus-stack) deploys and manages Prometheus on Kubernetes using CRDs: ServiceMonitor, PodMonitor, PrometheusRule, AlertmanagerConfig. It is the standard way to run Prometheus on K8s. The operator handles configuration, scaling, and lifecycle. Most K8s monitoring stacks use it.

Related Tools

Dep EOL Scanner

Check your dependencies for end-of-life packages.

Stack Health Scorecard

Get an overall health grade for your tech stack.

Release Index

Browse all tracked platforms and their release timelines.
Browse All Version History