Elastic
6.9/10Elastic is a search AI platform built on Elasticsearch. Engineering teams use it for vector and hybrid search, retrieval-augmented generation, AI agents, observability and security analytics.
Data last updated November 2026 · Our methodology
Overview
Elastic is the company behind Elasticsearch, Kibana and Logstash. Its platform combines full-text, vector and hybrid search with machine learning and generative AI tooling. Teams use it to build retrieval-augmented generation (RAG) applications over their own data. Agent Builder lets them create AI agents that answer questions using tools and indexed data. The Elastic AI Assistant works across the observability and security products, and the platform also offers anomaly detection, an AI Playground and the Elastic Workflows automation feature. It suits developer-led mid-market and enterprise teams that already store logs, documents or product data in Elasticsearch. It is not a fit for non-technical users who want a ready-made writing or chat tool. Elastic Cloud Hosted is priced by resources and has four subscription tiers, starting at about 99, 114, 131 and 184 US dollars per month for Standard, Gold, Platinum and Enterprise, based on Elastic's reference configuration of 120 GB of storage across two zones. Single sign-on (SSO) and the managed LLM for Agent Builder are part of the Enterprise tier. Elastic Cloud Serverless is billed by usage: VCU-hours for ingest, search and machine learning, plus storage and egress. The managed LLM costs 4.50 dollars per million input tokens and 21 dollars per million output tokens. You can deploy on AWS, Google Cloud, Azure or Alibaba Cloud, in a FedRAMP environment, or self-managed on your own infrastructure. The main limitation is that operating the platform is complex, and G2 reviewers say costs climb as data volumes grow.
Our Rating
Who Elastic is best for
Good fit
It is a developer-led platform for vector and hybrid search, RAG and Agent Builder, with API access.
The Enterprise tier adds SSO, the managed LLM and Agent Builder, and the platform carries a 99.95% uptime SLA, FedRAMP availability and self-managed licensing for large data estates.
Its observability and security analytics products serve infrastructure teams.
Mid-market engineering teams can start on usage-based Serverless, but they need in-house expertise to run it well.
Product teams build search and RAG features over their product data with it.
Less suited
It offers FedRAMP deployment and self-managed options.
Technical startups can begin with Serverless or the free self-managed core, but G2 reviewers report a steep learning curve and costs that grow with data.
Hosted plans start at about 99 dollars a month, and the platform is complex for businesses without engineers.
The vendor says it is not a fit for non-technical users, and it has a steep learning curve.
Pros & Cons
Pros
- G2 reviewers often praise fast search and aggregations on large datasets.
- Search, vector retrieval, agents and observability data share one data store, so AI features can use existing indexes.
- Flexible deployment covers AWS, Google Cloud, Azure, Alibaba Cloud, FedRAMP and self-managed infrastructure.
- Serverless charges per unit, with separate VCU rates for ingest, search and machine learning, so lighter workloads cost less.
- The free self-managed core gives teams a no-cost starting point.
Cons
- Cost is the most common complaint on G2, especially as data volumes and licensed features grow.
- Reviewers report a steep learning curve and say setup and management need considerable expertise.
- Key AI features such as Agent Builder with the managed LLM and SAML SSO sit in the top Enterprise tier on Hosted.
- Elastic's own documentation warns that not every third-party LLM works reliably with Agent Builder.
Key Features
Elastic Pricing
Cloud Hosted Standard
- Limited support
- AutoOps monitoring
- Core Elasticsearch
- Usage-based instance pricing
- Starting price assumes 120 GB of storage across 2 zones
Cloud Hosted Gold
- Reporting
- Watcher alerting
- Everything in Standard
- Multi-stack monitoring
- Third-party alerting actions
Cloud Hosted Platinum
- 99.95% uptime SLA
- Everything in Gold
- ELSER and e5 models
- Cross-cluster replication
- Advanced security features
- Anomaly detection and supervised machine learning
Cloud Hosted Enterprise
- GPU inference
- Elastic Workflows
- SAML single sign-on
- Searchable snapshots
- Everything in Platinum
- AI insights for security
- AI Assistant for observability
- Agent Builder with a managed LLM
Self-managed
- Free basic tier with community support
- Paid licences priced by node count and RAM
Serverless
- Automatic scaling
- Usage-based billing
- 10,000 Workflow executions included
- 1,000 Agent Builder executions included
Frequently Asked Questions
Yes. Elastic offers a free trial of Elastic Cloud with no commitment, but its pricing pages do not state the trial length.
On Elastic Cloud Hosted, the AI Assistant, Agent Builder with a managed LLM and Elastic Workflows are part of the Enterprise tier, which starts at about 184 dollars a month. Serverless includes 1,000 Agent Builder executions.
The Elastic Managed LLM costs 4.50 dollars per million input tokens and 21 dollars per million output tokens. Elastic Inference Service models start at 0.08 dollars per million tokens.
Yes. Elastic can be self-managed, with paid licences priced by node count and RAM. It also runs on AWS, Google Cloud, Azure, Alibaba Cloud and a FedRAMP environment.
Platinum and Enterprise subscriptions include a 99.95% monthly uptime SLA on both Hosted and Serverless.
Elastic Cloud SAML SSO is listed under the Enterprise tier on Hosted.
Similar Tools
See all alternatives
Amazon's public cloud platform, with over 200 pay-as-you-go services for compute, storage, databases, networking and AI, used by startups through large enterprises and governments.
Anthropic's organization-wide Claude plan adds chat, Claude Code, Cowork agents, connectors, and enterprise governance such as SCIM, audit logs, and a compliance API.
Fin is Intercom's AI customer service agent. It resolves support conversations across chat, email, voice and social, either inside Intercom or on top of an existing helpdesk.