At a glance
| Supported databases | PostgreSQL, MySQL, SQL Server, Oracle, Db2 for LUW, Snowflake, Redshift, BigQuery, Databricks, MongoDB, Salesforce and others |
| Data generation | Fabricate generates data from scratch; Structural can synthesize column values |
| Masking | Structural generators per column; Textual for free text and files |
| Referential integrity | Subsetting keeps related rows together; consistent generators keep join values aligned |
| CI/CD integration | Structural API; specific CI tool plugins not documented on the pages checked |
| Data volume or subsetting | Subsetting on most connectors; not on Databricks, BigQuery, Iceberg, files or Spark |
| Deployment | Tonic Cloud; self-hosted Docker or Kubernetes on Enterprise plans |
| Licensing | Structural by quote on source data size; Fabricate $0, $29 per month or Enterprise; Textual per 1,000 words |
Tonic.ai is a fit when developers need production-shaped but de-identified copies of relational databases and warehouses, and want a modern web app and API to produce them. Structural is quote-only and priced on source data size, so the cheap entry point is Fabricate, which has a free plan and a $29 per month plan for generating data from a schema.
Overview
Tonic.ai is a test data company with three products. Tonic Structural reads a source database, applies a generator to each sensitive column (masking, consistent replacement or synthesis), optionally subsets the data, and writes a de-identified copy to a destination. Tonic Fabricate generates relational data, free text and mock APIs from a description or schema, driven by an AI agent and metered in usage credits. Tonic Textual detects and redacts or synthesizes sensitive values in free text and files.
Structural runs on Tonic Cloud or self-hosted in Docker Compose or on Kubernetes with Tonic's Helm chart; the documentation asks for an x86 host with 8 vCPUs, 16 GB RAM (32 GB recommended) and 100 GB disk, and says ARM is not supported. Self-hosting is an Enterprise plan feature. Its connectors cover the main relational engines, cloud warehouses, MongoDB and Salesforce, although some (Databricks, BigQuery, Iceberg, files and Spark) do not support subsetting.
Key features
- Column-level generators in Structural that mask, consistently replace or synthesize values while keeping the output usable for tests
- Subsetting in Structural to cut a large source down to a smaller, related slice (not available on every connector)
- Privacy Hub and schema change review in Structural, flagging sensitive columns and new columns before a generation runs
- Structural API and an AI chat assistant (Structural Agent) alongside the web application
- Fabricate agent that builds synthetic relational data, mock APIs and exports from a prompt or schema
- Textual detection and redaction of sensitive entities in unstructured text, with custom detection models
- Tonic Cloud or self-hosted deployment (Docker Compose or Helm on Kubernetes) for Structural and Textual Enterprise
Integrations and ecosystem
Pros and cons
Pros
- Covers masking, subsetting and from-scratch synthesis in one vendor, with separate products for each job
- Broad connector list that spans transactional databases, warehouses, MongoDB and Salesforce
- Fabricate has a published free plan and a low published monthly price for small teams
- Self-hosted option for organisations that cannot send production data to a SaaS
Cons
- Structural, the core masking product, has no published prices and is priced by source data volume
- Self-hosting needs the Enterprise plan; Professional is Tonic Cloud only
- Subsetting is not available on several connectors, including Databricks and BigQuery
- Three products with three pricing models make total cost harder to estimate
Pricing
Tonic.ai prices its three products separately. Tonic Structural (de-identification and subsetting) has Professional and Enterprise plans, both custom priced by annual contract based on source data size (on-disk size excluding logs and indexes); Professional covers up to 10 TB, 10 users and 2 data source types on Tonic Cloud, and Enterprise adds unlimited sources and self-hosting. Tonic Fabricate (synthetic data from scratch) has a Free plan at $0 per month with $5 of monthly credits, a Plus plan at $29 per month with $25 of credits and pay-as-you-go overage, and a custom-priced Enterprise plan. Tonic Textual (unstructured text) is billed at a flat rate per 1,000 words processed (rate not published) or by Enterprise contract. No free tier is listed for Structural or Textual.
Pricing checked on the vendor's official page on 7 November 2026. Prices change; confirm before buying.
Who it is best for
Engineering teams that want de-identified, subsetted copies of production databases and warehouses for development and testing, with an API to automate refreshes.
Typical use cases
- Refreshing staging databases with masked production data before each release
- Giving offshore or contractor developers realistic data without exposing personal data
- Shrinking a multi-terabyte database to a related subset for local development
- Generating synthetic seed data and mock APIs for a new application before production data exists
- Redacting personal data from support tickets or documents before using them for AI training
Who should look elsewhere
- Teams that only need a quick file of fake rows for a demo, where a lighter generator is enough: see Mockaroo.
- Mainframe-heavy estates that need Db2 for z/OS, IMS or VSAM coverage, where IBM Optim or GenRocket document that support.
Verdict
Tonic.ai is a credible choice for developer-led test data programmes that mix relational databases and cloud warehouses, with Structural handling masking and subsetting and Fabricate covering synthetic data. Budget for a quote on Structural, check that your connectors support subsetting, and compare it with Delphix or K2view if you also need database virtualization or entity-based provisioning.
Frequently asked questions
What is the difference between Tonic Structural, Fabricate and Textual?
Structural de-identifies and subsets existing structured data from your databases. Fabricate generates new synthetic data, mock APIs and exports from scratch. Textual detects and redacts sensitive values in unstructured text and files. Each is priced separately.
Is Tonic.ai free?
Only Tonic Fabricate has a free plan ($0 per month with $5 of monthly credits). Tonic Structural is sold by custom quote based on source data size, and Tonic Textual is billed per 1,000 words processed or by Enterprise contract.
Can Tonic Structural run on premises?
Yes, on the Enterprise plan. Tonic documents self-hosted deployment with Docker Compose or on Kubernetes using its Helm chart, on x86 hosts. The Professional plan runs on Tonic Cloud only.
Did Tonic.ai buy Mockaroo?
Tonic.ai's April 2025 press release says it acquired Fabricate, a product built by Mockaroo, and that Mockaroo founder Mark Brocato joined Tonic to lead it. The release does not say Mockaroo itself was acquired, and mockaroo.com still carries a Mockaroo, LLC copyright.
Sources
- Tonic.ai pricing
- Tonic Structural data connector summary
- Deploying a self-hosted Structural instance
- Structural system requirements
- Tonic Structural release notes
- Tonic.ai acquires Fabricate (press release, April 2025)
Checked November 2026.
Listed in: Database Testing & Test Data.
How we research reviews and what star ratings mean: our editorial method.