Self-hosted observability tools in 2026
Understand what a system is doing from its telemetry: metrics, logs, traces, profiles and errors. This list has 96 observability tools. Of these, 71 are open source, 91 have a free plan and 96 can be self-hosted. Checked by hand on 30 September 2026.
- 1
AgentOps
Observability for AI agents: records agent sessions with LLM calls, tool use and costs so you can replay and debug runs; SDK and app are open source.
- 2
Apache SkyWalking
APM for distributed systems from the Apache Software Foundation: agents and service mesh probes collect traces, metrics and logs and draw service topology maps.
- 3
Arize Phoenix
Source-available LLM tracing and evaluation tool built on OpenTelemetry and OpenInference; run it locally, in a notebook or on your servers to inspect traces and score outputs.
- 4
AWS Distro for OpenTelemetry
AWS-supported distribution of the OpenTelemetry Collector and SDKs, preconfigured to send traces and metrics to CloudWatch, X-Ray, Amazon Managed Prometheus and partner backends.
- 5
Backtrace
Crash and error reporting from Sauce Labs for native apps, game engines, consoles and servers, including minidump and core dump processing.
- 6
Bindplane
Telemetry pipeline built on OpenTelemetry: manages fleets of collectors from one console and filters, samples and routes logs, metrics and traces to any backend. Cloud or self-hosted.
- 7
Boomerang
Open source JavaScript library, maintained by Akamai, that measures page load and user-perceived performance in real browsers and beacons the timings to a server you choose.
- 8
Braintrust
Evaluation and observability platform for AI products: logs production traces, runs evals and scorers, and offers a prompt playground and datasets.
- 9
Bugsink
Sentry-SDK-compatible error tracker built to self-host as a single container on modest hardware. Groups errors into issues and sends chat or webhook alerts.
- 10
Chronograf
InfluxData's web UI for InfluxDB 1.x: builds dashboards over InfluxQL queries and manages Kapacitor alert rules.
- 11
ClickStack
ClickHouse's open source observability stack built on HyperDX: logs, traces, metrics and session replays stored in ClickHouse and searched from one UI.
- 12
Confident AI
Evaluation and observability platform from the makers of the open source DeepEval library: LLM tracing, regression tests in CI, and red teaming.
- 13
Coroot
Open source observability that uses eBPF to build service maps and collect metrics, logs, traces and profiles without code changes, with root-cause hints.
- 14
Cortex
Horizontally scalable, multi-tenant long-term storage for Prometheus metrics; a CNCF project.
- 15
Cribl
Telemetry pipeline (Cribl Stream, Edge, Search and Lake) that filters, reshapes and routes logs between collectors and destinations such as Splunk or S3 to cut volume.
- 16
CrowdStrike Falcon LogScale
Index-free log store (formerly Humio) that compresses and searches large log volumes quickly; sold by CrowdStrike for security and operations logging.
- 17
Dozzle
Small web UI that streams and searches live Docker container logs across hosts, with no log storage. Optional Dozzle Cloud adds grouped alerts and log analysis.
- 18
Dynatrace
Enterprise observability platform that instruments hosts and apps automatically through its OneAgent, maps dependencies and points to likely root causes; also offered as self-managed Dynatrace Managed.
- 19
Elastic Observability
Logs, metrics, APM traces and synthetics built on Elasticsearch and Kibana, run yourself or on Elastic Cloud.
- 20
Elastic Stack (ELK)
Elasticsearch, Logstash, Kibana and Beats: full-text indexing and search of logs with Kibana dashboards and alerting. Self-host the free distribution or buy Elastic Cloud.
- 21
Errbit
Self-hosted Rails app that receives errors from Airbrake-compatible clients, groups them and emails or notifies your team.
- 22
Exceptionless
Open source .NET-first error, log and feature-usage reporting with a hosted service or free self-hosting. Groups events into stacks and notifies on new or regressed errors.
- 23
Fluent Bit
Lightweight C log and metrics processor from the Fluentd project, common as the node-level agent in Kubernetes. Ships to Loki, Elasticsearch, cloud logging and OTLP.
- 24
Fluentd
CNCF graduated log collector that reads, parses, buffers and forwards logs to hundreds of outputs through a plugin system. Written in Ruby and C.
- 25
Future AGI
Open source platform to trace, evaluate, simulate and guard LLM and agent apps, including an AI gateway.
- 26
Galileo
AI evaluation and observability platform with tracing, evaluation metrics and real-time guardrails for agents; hosted, VPC or on-prem.
- 27
GlitchTip
Open source error tracker that accepts events from Sentry SDKs, with simpler infrastructure than self-hosted Sentry. Also does basic uptime checks. Hosted or self-hosted.
- 28
gProfiler
Open source system-wide profiler from Intel (built by Granulate) that merges several language profilers into one flame graph per host; a local backend and UI are published separately.
- 29
Grafana
Open source dashboards and alerting over Prometheus, Loki, Elasticsearch, SQL databases and many other data sources.
- 30
Grafana Alloy
Grafana's OpenTelemetry Collector distribution with built-in Prometheus scraping and a pipeline configuration language; collects metrics, logs, traces and profiles. Replaces Grafana Agent.
- 31
Grafana Faro
Open source web SDK for frontend observability: captures page performance, Web Vitals, errors, logs and traces from browsers and sends them to Grafana Cloud or a self-run Alloy and Loki stack.
- 32
Grafana Loki
Log aggregation system that indexes only labels, not full log text, and stores chunks in object storage. Queried with LogQL from Grafana; hosted as part of Grafana Cloud.
- 33
Grafana Mimir
Horizontally scalable, multi-tenant long-term storage for Prometheus metrics from Grafana Labs, backed by object storage.
- 34
Grafana Pyroscope
Continuous profiling database that collects CPU, memory and other profiles from running applications so you can find hot code paths over time.
- 35
Grafana Tempo
Distributed tracing backend that stores traces cheaply in object storage and accepts OpenTelemetry, Jaeger and Zipkin data.
- 36
Graphite
Time-series metrics storage and graphing: Carbon receives metrics, Whisper stores them, and Graphite-web renders graphs and serves a query API.
- 37
Graylog
Centralised log collection with parsing pipelines, stream routing, search and alert conditions. Graylog Open is free to self-host; paid editions add security analytics and archiving.
- 38
groundcover
Kubernetes observability that uses an eBPF sensor and runs in your own cloud account, covering logs, traces, metrics and APM without code changes.
- 39
Helicone
Open source LLM proxy and logging: point your provider base URL at it to record every request with cost, latency and user, plus caching and rate limits.
- 40
Honeycomb Refinery
Open source tail-based sampling proxy for traces: buffers whole traces and keeps errors, slow requests and a sample of the rest before forwarding to Honeycomb.
- 41
HoneyHive
Observability and evaluation platform for AI agents: OpenTelemetry-based tracing, automated and human evals, prompt versioning and production monitoring.
- 42
IBM Instana
Enterprise APM and observability that discovers services automatically and traces every request across them.
- 43
InfluxDB
Time-series database for metrics and events; InfluxDB 3 Core is open source, with Enterprise and managed cloud editions.
- 44
Jaeger
CNCF distributed tracing platform for collecting, storing and exploring request traces across microservices; version 2 is built on the OpenTelemetry Collector.
- 45
Laminar
Open source observability for AI agents: OpenTelemetry-based tracing, evaluations and analysis of agent runs, self-hostable or cloud.
- 46
Langfuse
Open source LLM engineering platform: traces LLM and agent calls, manages prompts, runs evaluations and tracks cost and latency. Self-host or use the cloud; acquired by ClickHouse in 2026.
- 47
LangSmith
LangChain's platform for tracing, evaluating and monitoring LLM apps and agents, usable with or without LangChain; also hosts agent deployments.
- 48
LangWatch
Open source LLMOps platform: traces LLM and agent runs, runs evaluations and agent simulations, and helps optimize prompts; cloud or self-managed.
- 49
Latitude
Open source platform to trace, monitor and evaluate AI agents in production, with prompt management; self-host or use the hosted plans.
- 50
lnav
Terminal log file navigator that merges several log files into one timeline, highlights errors, and lets you query log lines with SQL. Runs locally, no server.
- 51
Logflare
Log ingestion and search backed by Google BigQuery, popular with Cloudflare Workers, Vercel and Elixir apps. Now part of Supabase and open source.
- 52
Nightingale
Open source alerting and dashboard platform that sits on Prometheus, VictoriaMetrics, Elasticsearch and other sources, with alert rules and notification routing.
- 53
NXLog
Log collection agent and platform strong on Windows event logs: collects, converts and forwards logs to SIEM and observability tools. A free Community Edition is also available.
- 54
Odigos
Uses eBPF to add OpenTelemetry tracing to applications on Kubernetes without code changes and sends the data to any backend.
- 55
OpenLIT
Open source, OpenTelemetry-native AI observability: SDK auto-instruments LLMs, vector databases and GPUs, with a self-hosted UI for traces, costs, evals and prompts.
- 56
OpenLLMetry
Open source OpenTelemetry instrumentations for LLM providers, vector databases and agent frameworks, by Traceloop; sends traces to any OTLP backend or the Traceloop platform.
- 57
OpenObserve
Single-binary store for logs, metrics and traces on object storage, with SQL search, dashboards and alerts. Self-host the open source build or use the pay-as-you-go cloud.
- 58
OpenReplay
Open source session replay with developer tools attached: console logs, network requests and performance data alongside each recording, plus co-browsing. Self-host free or use a managed instance.
- 59
OpenSearch
Apache-licensed fork of Elasticsearch and Kibana run under the Linux Foundation. Indexes and searches logs, with OpenSearch Dashboards, alerting and trace analytics plugins.
- 60
OpenTelemetry
CNCF project defining vendor-neutral APIs, SDKs, semantic conventions and the OTLP protocol for traces, metrics, logs and profiles, with auto-instrumentation for major languages.
- 61
OpenTelemetry Collector
Vendor-neutral agent and gateway that receives, processes and exports traces, metrics and logs to one or more backends.
- 62
OpenTelemetry eBPF Instrumentation (OBI)
Zero-code eBPF instrumentation that produces traces and RED metrics for HTTP, gRPC, database and messaging calls from Linux processes. Started as Grafana Beyla and was donated to OpenTelemetry.
- 63
OpenTelemetry eBPF Profiler
Whole-system eBPF profiling agent donated to OpenTelemetry by Elastic. Profiles native, JVM, Python, Ruby, PHP, Node.js and Go code without instrumentation and exports via OTLP.
- 64
Opik
Open source LLM and agent observability from Comet: tracing, test suites, evaluations and a prompt optimizer; self-host the same code as the cloud.
- 65
Parca
Open source continuous profiler: an eBPF agent samples CPU stacks from every process on a host with no code changes, and the server stores profiles for querying as flame graphs over time.
- 66
Parseable
Log and telemetry store written in Rust that keeps data as Parquet on object storage and queries it with SQL. Open source server, plus a hosted cloud and enterprise edition.
- 67
Perforator
Continuous profiler open sourced by Yandex for large fleets: deploys to Kubernetes, collects low-overhead eBPF profiles cluster-wide, and can feed profile-guided optimization builds.
- 68
Perses
CNCF open source dashboard tool with an open dashboard specification: Prometheus, Tempo and Loki panels, dashboards as code, and Kubernetes-native deployment.
- 69
Pinpoint
APM for large Java and PHP systems that traces transactions across services and draws a server map of their calls.
- 70
Pixie
Kubernetes observability that uses eBPF to capture service traffic, metrics and CPU profiles automatically; a CNCF sandbox project.
- 71
Portkey
AI gateway with an observability layer: one API for many model providers with routing, fallbacks and caching, and logs of each request with cost and latency. The gateway is open source.
- 72
PostHog
Open source product suite with session replay, product analytics, feature flags, error tracking and LLM analytics sharing one event store. Hosted with a monthly free allowance, or self-hosted.
- 73
Prometheus
CNCF metrics system that scrapes time series from exporters and apps, stores them locally, and evaluates PromQL alert rules that it sends to Alertmanager.
- 74
Quickwit
Rust search engine for logs and traces that indexes into object storage and searches there directly, with Elasticsearch-compatible and Jaeger APIs. The team joined Datadog in January 2025.
- 75
Respan
LLM gateway, logging and evaluation platform (formerly Keywords AI): routes calls to many models through one API and records traces, cost and scores.
- 76
rrweb
Open source library that records DOM changes in the browser and replays them as a session video; the recording engine inside many session replay products.
- 77
rsyslog
Default syslog daemon on many Linux distributions; also a high-throughput log pipeline that parses, filters and forwards logs to Elasticsearch, Kafka, files or remote syslog.
- 78
Sentry
Captures exceptions and crashes from web, mobile and backend SDKs, groups them into issues with stack traces and releases, and alerts owners. Adds tracing, replays and logs.
- 79
Seq
Self-hosted server for structured logs and traces, popular in .NET shops via Serilog. Search with a SQL-like syntax, build dashboards and send alerts.
- 80
SigNoz
Open source, OpenTelemetry-native APM with traces, metrics and logs stored in ClickHouse, self-hosted or as SigNoz Cloud.
- 81
Smashing
Ruby (Sinatra) framework for team wallboard dashboards built from widgets that pull data from scripts and APIs; community fork of Shopify's Dashing.
- 82
Splunk
Log indexing and search platform (Splunk Enterprise, self-hosted, or Splunk Cloud) with the SPL query language, dashboards and alerting. Widely used for security and IT operations.
- 83
Splunk AppDynamics
Enterprise APM that maps business transactions across services and ties application performance to business metrics; now part of Splunk and Cisco.
- 84
Superlog
Open source OpenTelemetry observability for traces, logs and metrics, with AI agents that investigate incidents; self-hostable or hosted.
- 85
syslog-ng
Long-running open source log collector and forwarder: receives syslog, files and journald, parses and filters messages, and sends them to files, databases, Kafka or SIEMs. Commercial editions exist.
- 86
Telebugs
Self-hosted error tracker that runs as one container and accepts reports from existing Sentry SDKs by changing the DSN. Sold as a licence, not a subscription.
- 87
Telegraf
InfluxData's plugin-based agent that collects metrics and events from systems, services and APIs and writes them to InfluxDB, Prometheus and many other outputs.
- 88
Tenzir
Open source pipeline engine for security and log data: nodes collect, parse, normalise and route events with a pipeline language, managed from a hosted or self-run platform.
- 89
Thanos
Adds a global query view, high availability and long-term object-storage retention to existing Prometheus servers; a CNCF project.
- 90
Uptrace
OpenTelemetry APM that stores traces, metrics and logs in ClickHouse, available self-hosted or as a cloud service priced per GB.
- 91
Vector
Rust observability pipeline from Datadog that collects, transforms and routes logs and metrics between sources and sinks, run as agent or aggregator.
- 92
VictoriaLogs
Log database from the VictoriaMetrics team, run as a single binary or cluster, queried with LogsQL and viewable in Grafana. Accepts Loki, Elasticsearch and OTLP ingestion.
- 93
VictoriaMetrics
Prometheus-compatible time-series database and monitoring stack in single-node and cluster versions, with enterprise and managed cloud editions.
- 94
W&B Weave
Weights & Biases toolkit for tracing LLM calls, running evaluations with custom scorers and comparing model and prompt versions.
- 95
Zipkin
Distributed tracing system that collects timing data from instrumented services and shows where latency builds up across them.
- 96
zymtrace
Self-hosted, cluster-wide profiler for AI training and inference fleets: continuous whole-system CPU profiling plus GPU and accelerator profiling for CUDA, ROCm and other chips.
Side by side
| Tool | Pricing | Free plan | Open source | Self-hosted |
|---|---|---|---|---|
| AgentOps | Free plan | Yes | Yes | Yes |
| Apache SkyWalking | Free | Yes | Yes | Yes |
| Arize Phoenix | Free | Yes | No | Yes |
| AWS Distro for OpenTelemetry | Free | Yes | Yes | Yes |
| Backtrace | Free plan | 25K errors/month, 1 month retention, 10 GB | No | Yes |
| Bindplane | Free plan | 10 collectors, 100 GB/day, 1 project | No | Yes |
| Boomerang | Free | Yes | Yes | Yes |
| Braintrust | Free plan | 1 GB processed data, 10k scores, 14-day retention | No | Yes |
| Bugsink | Free plan | Hosted: 15K events/month; self-hosted free | No | Yes |
| Chronograf | Free | Yes | Yes | Yes |
| ClickStack | Free plan | Yes | Yes | Yes |
| Confident AI | Free plan | 2 seats, 1 project, 1 GB-month of trace spans | No | Yes |
| Coroot | Free plan | Yes | Yes | Yes |
| Cortex | Free | Yes | Yes | Yes |
| Cribl | Free plan | Up to 1 TB/day | No | Yes |
| CrowdStrike Falcon LogScale | Paid | No | No | Yes |
| Dozzle | Free plan | Yes | Yes | Yes |
| Dynatrace | Paid | No | No | Yes |
| Elastic Observability | Free plan | Yes | Yes | Yes |
| Elastic Stack (ELK) | Free plan | Yes | Yes | Yes |
| Errbit | Free | Yes | Yes | Yes |
| Exceptionless | Free plan | 3,000 events/month, 1 user, 3-day retention | Yes | Yes |
| Fluent Bit | Free | Yes | Yes | Yes |
| Fluentd | Free | Yes | Yes | Yes |
| Future AGI | Free plan | Yes | Yes | Yes |
| Galileo | Free plan | Yes | No | Yes |
| GlitchTip | Free plan | 1,000 events/month | Yes | Yes |
| gProfiler | Free | Yes | Yes | Yes |
| Grafana | Free plan | Yes | Yes | Yes |
| Grafana Alloy | Free | Yes | Yes | Yes |
| Grafana Faro | Free plan | Yes | Yes | Yes |
| Grafana Loki | Free plan | Grafana Cloud: 50 GB logs/month, 14-day retention | Yes | Yes |
| Grafana Mimir | Free plan | Yes | Yes | Yes |
| Grafana Pyroscope | Free plan | Yes | Yes | Yes |
| Grafana Tempo | Free plan | Yes | Yes | Yes |
| Graphite | Free | Yes | Yes | Yes |
| Graylog | Free plan | Yes | No | Yes |
| groundcover | Free plan | Yes | No | Yes |
| Helicone | Free plan | 10,000 requests, 1 GB storage, 1 seat | Yes | Yes |
| Honeycomb Refinery | Free | Yes | Yes | Yes |
| HoneyHive | Free plan | 10,000 events/mo, 5 users, 30-day retention | No | Yes |
| IBM Instana | Paid | No | No | Yes |
| InfluxDB | Free plan | Yes | Yes | Yes |
| Jaeger | Free | Yes | Yes | Yes |
| Laminar | Free plan | 1 GB data, 7-day retention, 1 seat | Yes | Yes |
| Langfuse | Free plan | 50k units/mo, 30 days data access, 2 users | Yes | Yes |
| LangSmith | Free plan | 5k base traces/mo, 1 seat | No | Yes |
| LangWatch | Free plan | 50k events/mo, 14-day data access, 2 users | Yes | Yes |
| Latitude | Free plan | 20K credits/mo, 30-day retention | Yes | Yes |
| lnav | Free | Yes | Yes | Yes |
| Logflare | Free plan | 12.96M events/month, 3-day retention | Yes | Yes |
| Nightingale | Free plan | Yes | Yes | Yes |
| NXLog | Free plan | Yes | No | Yes |
| Odigos | Free plan | Yes | Yes | Yes |
| OpenLIT | Free | Yes | Yes | Yes |
| OpenLLMetry | Free plan | Traceloop: 50K spans/mo, 5 seats, 24h retention | Yes | Yes |
| OpenObserve | Free plan | Yes | Yes | Yes |
| OpenReplay | Free plan | Yes | Yes | Yes |
| OpenSearch | Free | Yes | Yes | Yes |
| OpenTelemetry | Free | Yes | Yes | Yes |
| OpenTelemetry Collector | Free | Yes | Yes | Yes |
| OpenTelemetry eBPF Instrumentation (OBI) | Free | Yes | Yes | Yes |
| OpenTelemetry eBPF Profiler | Free | Yes | Yes | Yes |
| Opik | Free plan | 25k spans/mo, 10 members, 60-day retention | Yes | Yes |
| Parca | Free | Yes | Yes | Yes |
| Parseable | Free plan | Yes | Yes | Yes |
| Perforator | Free | Yes | Yes | Yes |
| Perses | Free | Yes | Yes | Yes |
| Pinpoint | Free | Yes | Yes | Yes |
| Pixie | Free | Yes | Yes | Yes |
| Portkey | Free plan | 10k logs/mo, 3-day log retention | No | Yes |
| PostHog | Free plan | 5K session recordings and 1M events per month | Yes | Yes |
| Prometheus | Free | Yes | Yes | Yes |
| Quickwit | Free | Yes | Yes | Yes |
| Respan | Free plan | 100k logs, 1k scores | No | Yes |
| rrweb | Free | Yes | Yes | Yes |
| rsyslog | Free | Yes | Yes | Yes |
| Sentry | Free plan | Developer: 5K errors/month, 1 user | No | Yes |
| Seq | Free plan | Individual: 1 user, 50 GB storage | No | Yes |
| SigNoz | Free plan | Yes | Yes | Yes |
| Smashing | Free | Yes | Yes | Yes |
| Splunk | Free plan | Splunk Free: 500 MB/day, one instance, no alerting | No | Yes |
| Splunk AppDynamics | Paid | No | No | Yes |
| Superlog | Free plan | Yes | Yes | Yes |
| syslog-ng | Free plan | Yes | Yes | Yes |
| Telebugs | Paid | No | No | Yes |
| Telegraf | Free | Yes | Yes | Yes |
| Tenzir | Free plan | Community edition: 1 TB/day ingress | Yes | Yes |
| Thanos | Free | Yes | Yes | Yes |
| Uptrace | Free plan | 50 GB/mo | Yes | Yes |
| Vector | Free | Yes | Yes | Yes |
| VictoriaLogs | Free plan | Yes | Yes | Yes |
| VictoriaMetrics | Free plan | Yes | Yes | Yes |
| W&B Weave | Free plan | Yes | No | Yes |
| Zipkin | Free | Yes | Yes | Yes |
| zymtrace | Free plan | CPU profiling on up to 5 machines | No | Yes |
Plans and prices change; check each tool's own site before you decide.