Where does my data live and who owns it?

Telemetry is stored in open Parquet format on object storage — you can bring your own bucket or use BYOC/self-hosted deployments for complete data ownership, with data residency and sovereignty options on Enterprise.

What security certifications does Parseable have?

Parseable is GDPR certified and SOC 2 Type II compliant.

Parseable — Unified, AI-native observability on object storage

Parseable is an open-core observability platform that unifies logs, metrics, and traces in a single system built on a columnar (Parquet) data lake architecture. It is written in Rust, stores telemetry in open formats on S3-compatible object storage, supports SQL and PromQL queries, and includes AI-native features such as natural-language Q&A (Keystone), anomaly detection, forecasting alerts, and AI-assisted root cause analysis. It can be self-hosted as OSS, run as a managed cloud, or deployed in enterprise configurations with BYOC and data sovereignty options.

Key Features

Unified telemetry

Logs, metrics, and distributed traces in one platform, with views for alerts, dashboards, service maps, and an error page, all built into a single binary.

SQL and PromQL querying

Query telemetry with SQL, plus PromQL support and dashboards. Natural-language querying is available via the Keystone AI interface.

Object-store-native columnar storage

Telemetry is stored in open Parquet format on object storage (bring your own bucket), with a stateless design, smart caching, and high availability clustering for scale.

AI-native analysis

LLM and ML-powered capabilities including anomaly detection alerts, forecasting alerts, AI SQL generation, summarization, and AI-assisted root cause analysis across logs, metrics, traces, and events.

APM and AI agent observability

Automatic error tracking and APM with a service view, including observability for AI agents themselves.

Alerts and access control

Threshold and PromQL-based alerts, dashboards, OpenID Connect SSO, role-based access control, unlimited users, and (on Enterprise) audit logging and multi-tenant clusters.

OpenTelemetry-native with 60+ integrations

Integrates with Prometheus, Grafana, Kafka, Kubernetes, Fluent Bit, Fluentd, Vector, AWS, Azure, OpenAI, Anthropic, and more.

Companion tooling

MCP server for connecting AI agents to live telemetry, a Slack bot for investigations, PAI for automatic Kubernetes instrumentation, PB CLI for terminal-based querying, and OTex for AI-assisted instrumentation planning.

How It Works

Deploy

Install via a one-line curl setup script (curl -fsSL https://logg.ing/setup | bash), download the OSS binary from GitHub, or use managed Parseable Cloud / Enterprise deployment in public cloud, private cloud, hybrid, or BYOC.

Point at an object store and configure agents

Deploy on public or private cloud, point Parseable at an object store, and configure your telemetry agents (OpenTelemetry-native, plus Fluent Bit, Vector, Fluentd, eBPF, etc.).

Query, alert, and investigate

Use SQL or natural language to query full-fidelity data stored in open formats, build dashboards, set alerts, and let AI features surface anomalies and root causes.

Pros & Cons

Pros

Significantly lower cost than incumbent observability platforms (claims up to 80x lower cost than Datadog and Splunk) with predictable ingestion-based pricing ($0.39/GB)
Truly open core: AGPLv3 OSS version with SQL queries, dashboards, alerts, SSO, and unlimited users
Open Parquet storage on your own object store means no vendor lock-in and full data ownership
Single binary covering logs, metrics, traces, dashboards, and alerts simplifies the stack
AI features (anomaly detection, forecasting, natural-language Q&A, root cause analysis) are built in rather than bolted on

Who It's For

Best for

Engineering teams that want logs, metrics, and traces in one platform without Datadog/Splunk-scale bills
Organizations with strict data sovereignty, residency, or compliance requirements that want telemetry in their own object storage
Teams already using OpenTelemetry, Prometheus, Grafana, Fluent Bit, or Vector who want an OTel-native backend
Platform teams that prefer open formats (Parquet) and self-hosted or BYOC deployment
Fast-growing teams that want AI-assisted anomaly detection and root cause analysis out of the box

Not ideal for

Teams that need a zero-cost production option with clustering and RBAC groups — those features are only in paid tiers

Use Cases

AI-assisted anomaly investigation (e.g., an agent detects an error-rate spike, correlates logs, metrics, traces, and events, and identifies root cause)

High-cardinality telemetry handled via columnar design

Cost-efficient large-scale telemetry storage on object storage

Data sovereignty and compliance-sensitive deployments (SOC 2 Type II, GDPR certified)

Unified observability replacing separate log/metric/trace tools

Pricing

Ingestion-based pricing with three tiers: a free AGPLv3 OSS self-hosted version, Parseable Cloud at $0.39 per GB ingested (14-day free trial, 365-day retention, 99.999% uptime), and Enterprise starting at $15,000/year with dedicated SLA, flexible deployment (self-hosted, private cloud, hybrid, BYOC), and data sovereignty options. The site claims up to 80% lower telemetry costs versus Datadog and S

OSS

Self-hosted, AGPLv3 licensed. Includes SQL query and dashboards, threshold alerts, OIDC SSO, unlimited users, customizable retention, and self-managed object storage. Limited to 1 query node; RBAC groups not supported.

Cloud

$0.39 per GB ingested

14-day free trial, pay as you go. Includes 365-day retention, 99.999% uptime, AI-native analysis, anomaly detection, PromQL, unlimited users, dashboards, alerts, API access, and 24x7 support via portal. Shared dedicated deployment environme

Enterprise

Starts at $15,000/year

Everything in Cloud plus premium support with dedicated SLA, flexible deployment (self-hosted, private cloud, hybrid, BYOC), custom partitioning, multi-tenant cluster, audit logging, data residency & sovereignty options, and unlimited reten

FAQ

What are the key features of Parseable?

Parseable unifies logs, metrics, and traces in one open observability platform. Teams get SQL and natural-language querying, dashboards, alerts, access controls, forecasting, and object-storage-native data management.

Is Parseable open source?

Yes. The OSS version is AGPLv3 licensed and free to self-host, with SQL queries, dashboards, threshold alerts, OIDC SSO, and unlimited users. Paid Cloud and Enterprise tiers add scale, AI, and support features.

How is Parseable priced?

OSS is free. Parseable Cloud costs $0.39 per GB ingested (pay as you go, with a 14-day free trial). Enterprise starts at $15,000/year. The company claims up to 80% lower cost than Datadog and Splunk.

Can I use my existing observability stack?

Parseable