OpenObserve
Open-source observability platform with built-in AI observability and an AI SRE agent.
About
OpenObserve is an open-source, Rust-built observability platform (AGPL-3.0 for self-hosting) that unifies logs, metrics, traces, and real user monitoring in a single storage and query layer, positioning itself as a high-performance, low-cost alternative to Datadog, Splunk, Elasticsearch, and Grafana. It queries with SQL and PromQL, ingests OpenTelemetry natively, and the company reports roughly 140 times lower storage costs than Elasticsearch-based stacks on typical log workloads.
With its v1.0 release in September 2026, OpenObserve added AI Observability alongside classic telemetry: token cost and usage tracking across more than 80 providers and frameworks, full-session tracing of agent conversations including tool calls and per-step latency, agent behavior graphs, and built-in evaluation with custom scorers and LLM-as-judge. An O2 AI SRE Agent automates root cause analysis across the full telemetry set, and an Insights feature surfaces likely incident causes in under a minute.
What it can do
- Unified logs, metrics, traces, RUM, and AI observability in one platform
- Token cost and usage tracking across 80 plus AI providers and frameworks
- Agent session tracing with behavior graphs and evaluation tooling
- O2 AI SRE Agent for automated root cause analysis
- SQL and PromQL querying over columnar storage with usage-based pricing
Setup
Self-host the single binary or sign up for OpenObserve Cloud; instrument apps with OpenTelemetry.
Prerequisites
Telemetry sources to ingest; OpenTelemetry instrumentation for AI agent tracing.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 125; Trending score 80/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.