Coralogix
by Coralogix
AI-native observability that queries logs, metrics and traces in place — index-free, in your own S3 bucket
Coralogix is a full-stack, AI-native observability platform for SRE, platform and security engineering teams that analyses logs, metrics and traces in-stream and then queries them index-free directly inside the customer's own object storage. It removes the indexing and per-host costs that make traditional observability unaffordable at petabyte scale, and includes every enterprise control at every spend level.
Coralogix is a full-stack observability platform that ingests logs, metrics, traces and real-user-monitoring data over OpenTelemetry and analyses it in-stream, before it is written anywhere. Its Streama architecture applies alerting, enrichment, parsing and anomaly detection to telemetry as it flows, then parks the raw data in the customer's own S3 bucket, where Coralogix queries it remotely and index-free using DataPrime, its query language. That design removes the index-and-store cost curve that makes per-host or per-GB-indexed platforms expensive at scale, and it is why both the vendor and its users on Hacker News describe roughly 40-70% cost reductions for teams migrating off Datadog. Pricing is metered purely by ingested volume — $0.42/GB for logs, $0.16/GB for traces, $0.05/GB for metrics (1GB = 1,000 time series) and $1.50 per million AI tokens — with unlimited users, hosts, sources and team members, and every enterprise control (RBAC, SSO, SAML, audit trails, IP access control) included at every spend level rather than gated behind a premium tier. On the same data foundation Coralogix layers AI-native tooling: Olly, a built-in conversational AI agent; an MCP server and a programmable agentic CLI for automated workflows; and AI Discovery, AI Guardrails and LLM session and cost tracking for teams running generative-AI applications in production. The platform also covers APM, RUM, infrastructure monitoring, log analytics and SIEM/MDR security tooling, with 300+ integrations and native support for OpenTelemetry and Prometheus. Founded in 2014 by Ariel Assaraf and Yoni Farin, Coralogix raised a $200M Series F on 3 June 2026 co-led by Advent, CPPIB and Greenfield at a $1.6B valuation, bringing total funding to $550M, and serves more than 5,000 customers including IBM, Adobe, Monday.com, Puma, Palo Alto and JFrog across eight regions including AWS GovCloud.
The VP of Engineering or head of platform/SRE at a company whose observability bill is growing faster than its infrastructure, and who needs petabyte-scale log retention without an indexing tax or per-host licensing.
Keep full-fidelity telemetry with infinite retention in your own cloud bucket and pay only for ingested volume — typically 40-70% less than an indexed, per-host platform.
At a Glance
- Category
- Infrastructure & Cloud
- Pricing
- Usage-based
- Target Market
- CTOs, VPs of Engineering, SRE and Platform Teams, Enterprise Developers, CISOs
- Deployment
- Cloud-first, Multi-cloud, Hybrid
- Founded
- 2014
- Team Size
- 500+
- Customers
- 5,000+ customers worldwide (June 2026 funding announcement); the vendor's about page cites 4,000+ teams
Key Features
- ✓Streama in-stream analysis
Alerts, parses and detects anomalies on telemetry as it flows, before any of it is indexed or stored, so detection is not gated on storage cost.
- ✓Index-free remote query on your own bucket
Raw data lands in the customer's S3 bucket with infinite retention and Coralogix queries it in place via DataPrime, removing the indexing tax entirely.
- ✓Olly conversational AI agent
A built-in AI agent that answers questions about the running system in natural language across logs, metrics and traces on the same data foundation.
- ✓MCP server and agentic CLI
Exposes the platform to Model Context Protocol clients and scriptable agents so automated workflows can investigate incidents without a human in the console.
- ✓AI observability with guardrails
AI Discovery, AI Guardrails and an LLM session explorer track prompt cost and behaviour for generative-AI applications running in production.
- ✓Unified APM, RUM, infrastructure and SIEM/MDR
Application performance, real-user monitoring, infrastructure metrics and security monitoring all run on one pipeline with 300+ integrations.
- ✓No feature gating by tier
RBAC, SSO, SAML, audit trails, IP access control and unlimited users are included at every spend level rather than sold as an enterprise upgrade.
Capabilities
Use Cases
- •Cutting an observability bill without cutting visibility
Teams migrating from an indexed, per-host platform keep full log fidelity and infinite retention while paying only for ingested gigabytes.
- •Incident investigation across a full year of history
Engineers query a year or more of raw logs held in their own bucket during a postmortem, without rehydrating archived data first.
- •Monitoring generative-AI applications in production
Platform teams track LLM token spend, prompt sessions and guardrail violations alongside the infrastructure telemetry from the same application.
- •Running security monitoring on engineering telemetry
Security teams operate SIEM and managed detection on the same ingested data instead of duplicating the pipeline into a separate SIEM vendor.
- •Agent-assisted operations via MCP
An AI coding or ops agent connects through the MCP server to pull live system state and triage an alert before an engineer is paged.
Ideal For
Best For
- ✓Migrating off Datadog, Splunk or Elastic when log volume, not headcount, is driving the bill
- ✓Petabyte-scale log retention where data must stay in the customer's own S3 bucket for cost or sovereignty reasons
- ✓OpenTelemetry-standardised estates that want logs, metrics, traces, APM and RUM on one backend
- ✓Regulated teams needing SOC 2 Type II, ISO 27001/42001, PCI DSS, HIPAA and DORA alignment on an observability vendor
- ✓Security teams that want SIEM and MDR running on the same telemetry pipeline as engineering observability
Not Ideal For
- ✗Small teams with modest telemetry volume — the units-based model and DataPrime query language add operational overhead that a simple hosted log viewer avoids
- ✗Buyers whose primary requirement is deep synthetic monitoring; at least one Hacker News commenter chose Datadog specifically because Coralogix's synthetics did not cover their needs
- ✗Teams that need an on-premise, air-gapped deployment — Coralogix is a multi-tenant cloud SaaS, and only the raw data storage lives in your bucket
- ✗Organisations that want a fixed, predictable annual seat price rather than a bill that moves with ingested gigabytes
Integrations
Deployment
Market & Ratings
5,000+ customers worldwide (June 2026 funding announcement); the vendor's about page cites 4,000+ teams
Market Analysis
Pros
- ✓Cost model decouples retention from price — PeerSpot and Hacker News users independently describe large savings versus Datadog and Splunk at high log volume
- ✓Full OpenTelemetry and Prometheus compatibility with 300+ integrations, so migration does not require re-instrumentation
- ✓AI-powered anomaly detection and centralised log management measurably cut triage time; one PeerSpot reviewer reported resolution dropping from one to two hours down to twenty to thirty minutes
- ✓Unusually broad compliance coverage for an observability vendor: SOC 2 Type II, ISO 27001/27701/27017/27018/42001, PCI DSS v4.0.1, HIPAA, GDPR and DORA
- ✓Migration support from competing platforms is repeatedly praised in reviews
Cons
- ✗Cost is the single most-cited complaint on PeerSpot despite the cheaper model — one reviewer asked directly for licence cost reductions in high-volume environments
- ✗Complex queries over very large datasets are slow; a PeerSpot reviewer noted 'some complex queries take time to return results' at very high log volumes
- ✗The UI draws mixed feedback — one engineer called it 'too cluttered', and reviewers report a steep learning curve on advanced features and DataPrime
- ✗Feature gaps remain in synthetic monitoring; a Hacker News prospect chose Datadog because Coralogix's synthetics were insufficient at the time
- ✗Documentation and onboarding are cited as needing simplified UI options and guided templates
- ✗Practitioner discussion on Hacker News is thin — the two indexed submissions drew one point and zero comments each, so independent community signal is limited relative to Datadog or Grafana
Pricing
Free Trial
$0
- ✓14 days
- ✓No credit card required
- ✓Full feature access
- ✓Quota of 8 units
Units-based platform (all features included)
From $0.05 per GB
- ✓Logs $0.42/GB
- ✓Traces $0.16/GB
- ✓Metrics $0.05/GB (1GB = 1,000 time series)
- ✓AI $1.50 per 1M tokens
- ✓Infinite retention stored in your bucket
- ✓Unlimited users, hosts and sources
- ✓RBAC, SSO, SAML, audit trails
- ✓24/7 support from software engineers
List pricing is published and metered purely by ingested volume, not by seat or host: one unit equals $1.50 of data, logs bill at $0.42/GB, traces $0.16/GB, metrics $0.05/GB and AI at $1.50 per million tokens, with daily quota renewal. Unlimited users, hosts, sources and team members are included, and Coralogix states no features are locked behind an enterprise tier — RBAC, SSO, SAML, audit trails and compliance controls ship at every spend level. A 14-day free trial with an 8-unit quota needs no credit card. PeerSpot reviewers nonetheless flag licence cost as the top improvement area in very high-volume environments, so model your real ingest before committing.
Security & Compliance
Connect
Sources
This page was written from 7 sources, 3 on domains other than coralogix.com.
Stay Ahead of the Curve
Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.
SubscribeRelated Products
OpenObserve
Open-source, Rust-based observability on object storage — logs, metrics, traces and LLM telemetry in one binary
SkyPilot
Run and scale AI workloads across Kubernetes, Slurm and 25+ clouds from one interface
MyDecisive
Open-source, OpenTelemetry-native telemetry control plane that acts on incidents instead of just charting them
General Compute
ASIC-based inference cloud built for agents, where latency compounds