Skip to the content
Know your tech before you buy it. Makers: get listed from $49/mo

FakerForge

fakerforge.com

  • Free plan
  • Windows
  • macOS
  • Linux
  • Web
  • API
  • Self-hosted
8.5 Club score Visit fakerforge.com

What FakerForge is

FakerForge generates synthetic datasets, seeds mock databases, and publishes mocked API endpoints for development and testing. Users can provide SQL schemas or migration files, choose row counts and field rules, and create records that preserve schema relationships. Supported databases or destinations include MySQL, PostgreSQL, SQLite, MariaDB, MongoDB, and BigQuery; output formats include SQL, JSON, CSV, and XML. Mock endpoints can return configurable fake JSON for success cases, errors, edge cases, or timeouts. Its open-source CLI can inspect local or private PostgreSQL and MySQL databases and put generated data into their tables. The CLI has installation options for Linux, macOS, and Windows on amd64 and arm64. FakerForge says CLI credentials remain in the operating system keychain and only schema shape is uploaded. It describes generated records as synthetic, without real names, email addresses, or phone numbers. A free plan is available; BASIC costs $10 a month, billed per month, and PRO costs $20 a month, billed per month. Limits apply to generation, API, and mocking actions, with high request volume potentially throttled or rejected.

Who it suits

FakerForge suits developers, QA engineers, data analysts, and product teams creating test data or mocked APIs. Its CLI is relevant to users working with local or private PostgreSQL and MySQL databases.

Good to know, and what to watch

  • Generates records that respect schema relationships.
  • Exports SQL, JSON, CSV, and XML.
  • Mocks success, error, edge-case, and timeout scenarios.
  • CLI supports Linux, macOS, and Windows.
  • Free plan is available.
  • Generation and API actions can be capped by plan.
  • High request volume may be throttled or rejected.
  • Paid BASIC starts at $10 a month.

Free

Free

Basic

$10/mo

Pro

$20/mo

Free plan

Getting started

  1. Open the FakerForge website
  2. Upload an SQL schema or migration file
  3. Set row counts and field rules
  4. Generate records that follow schema relationships
  5. Export data as SQL, JSON, CSV, or XML
  6. Install the open-source CLI on Linux, macOS, or Windows when needed

Limits to check

The FREE plan costs nothing and includes 1 database, 100 rows per table, 10 tables per database, 500 API requests, 1 mock API, 5 endpoints per mock, 2 scenarios per endpoint, and 500 mock requests. Generation, API, and mocking actions are capped by plan limits, and high request volume may be throttled or rejected.

The Club's review

FakerForge combines synthetic-data generation, database seeding, and API mocking in a web service and open-source CLI. It suits developers, QA engineers, analysts, and product teams that need structured test data alongside controllable API responses. Its breadth is useful, but quotas and throttling make volume planning essential.

Overview

Upload a SQL schema or migration file, set row counts and field rules, and generate records that preserve schema relationships. That makes FakerForge a better fit for populating structured test databases than for producing isolated sample records. Database seeding and API mocking sit alongside generation, so teams can prepare data and simulate API behavior in one workflow.

FakerForge describes its generated records as entirely synthetic, without real names, email addresses, or phone numbers. That is a useful safeguard for test datasets, though it does not remove the need to review how generated data is handled in a particular workflow.

Key features

  • Schema-aware generation: Row counts and field rules can be applied to uploaded schemas or migrations, with relationships respected. This is useful when test cases depend on connected records; table and row ceilings determine how much a plan can cover.
  • Database support and exports: Supported databases or destinations include MySQL, PostgreSQL, SQLite, MariaDB, MongoDB, and BigQuery. SQL, JSON, CSV, and XML exports give teams several ways to move generated data into their workflows.
  • API mocking: Fake JSON payloads can be configured for success states, errors, edge cases, and timeouts. That gives QA teams more control over response scenarios, but mock endpoints and requests are quota-bound.
  • Open-source CLI: The CLI introspects local or private PostgreSQL and MySQL databases and can push generated data into their tables. Credentials stay in the operating system keychain, while only schema shape is uploaded. This is a practical option for teams working with private databases, although the CLI's database support is narrower than the service's broader destination list.
  • MCP integrations and automation: The MCP server works with GitHub Copilot, Claude, Cursor, Windsurf, and other compatible clients. Data refresh automation and API access are also supported, extending use beyond one-off generation.

Generation, API, and mocking actions can be capped by plan limits; high request volumes may be throttled or rejected. Teams with unpredictable traffic should account for that rather than assume the service will absorb bursts.

Pricing

The free plan costs nothing, billed per month. It includes 1 database, 50,000 AI tokens, up to 100 rows per table and 10 tables per database, 500 API requests, 1 mock API, 5 endpoints per mock, 2 scenarios per endpoint, and 500 mock requests. Those limits are enough to explore the workflow or support a small test project, but they can constrain repeated or broader QA runs.

Basic costs $10 a month, billed per month. It raises the allowance to 5 databases, 1,000,000 AI tokens, 1,000 rows per table and 20 tables per database, 10,000 API requests, 10 mock APIs, 50 endpoints per mock, 10 scenarios per endpoint, and 10,000 mock requests. This is the sensible step up for a small team that has outgrown the free caps.

Pro costs $20 a month, billed per month. It includes 20 databases, 5,000,000 AI tokens, 100,000 rows per table and 50 tables per database, 50,000 API requests, 50 mock APIs, 100 endpoints per mock, 50 scenarios per endpoint, and 100,000 mock requests. The higher row and request limits make it a better fit for larger test workloads, though published quotas still matter for sustained high-volume use.

Support for integration help, enterprise licensing, or product questions is available by email. Contact FakerForge for custom pricing if enterprise licensing is the requirement.

Platforms

FakerForge is available as a web service and API, with the CLI supporting Linux, macOS, and Windows on amd64 and arm64. This gives teams both a hosted workflow and a command-line option for supported local or private PostgreSQL and MySQL databases.

Who it's for

FakerForge is a strong match for developers and QA engineers who need database-shaped test records and mock responses with varied outcomes. Data analysts and product teams can also use it to prepare synthetic datasets, while the CLI particularly suits teams seeding local or private PostgreSQL and MySQL environments. It is less compelling for projects whose data volumes or request patterns regularly exceed plan quotas.

Pros and cons

  • Pro: Schema-aware generation preserves relationships, making the output more useful for realistic database testing than disconnected sample rows.
  • Pro: Database seeding and configurable API scenarios are combined with generation, reducing the need to split these tasks across separate workflows.
  • Pro: The CLI can work with local or private PostgreSQL and MySQL databases while keeping credentials in the operating system keychain.
  • Con: Free-plan caps are modest for sustained use: 100 rows per table, one mock API, and 500 API and mock requests each.
  • Con: High request volume may be throttled or rejected, so teams with variable traffic need to plan around quotas.
  • Con: The CLI's direct database introspection covers PostgreSQL and MySQL, not every database or destination supported by the service.

Alternatives

Synthehol Dataset is worth considering for a free hosted or API option with 50K rows per month, five projects, five datasets, one user, and 30-day retention; its free allowance is organized around monthly credits and storage rather than FakerForge's per-table and mock quotas.

SimpleTest is the closer alternative when the priority is running tests: its free Starter plan allows 50 parallel runs and 500 execution minutes per month, with AI-generated datasets limited to one per test.

Synth Studio may suit teams seeking a free-to-use option that supports self-hosting and offers up to 1M rows per generation under an MIT license.

NVIDIA ShadowPlay is another free option, available for Windows with a supported GPU and the required GeForce driver.

SynthEngyne offers monthly free credits, with separate credit costs for rows and images; consider it if that credit-based approach better fits your workload.

Anyverse is a paid web option with a free trial.

MOSTLY AI offers a free plan and supports API, Linux, self-hosted, and web use.

REaLTabFormer is a free, MIT-licensed option for Linux, macOS, Windows, or self-hosted use, built for Python 3.7 or later.

For a broader comparison, browse AI Synthetic Data Generators or Test Data Management Software.

Verdict

Choose FakerForge if your team wants schema-aware synthetic records, database seeding, and configurable API mocks in one toolkit, especially with the CLI for local or private PostgreSQL and MySQL databases. Its main drawback is quota dependence: teams with larger or erratic workloads should compare limits carefully or look for an option better suited to their volume.

The verdict

FakerForge covers synthetic data generation, database seeding, and API mocking, with both a web service and an open-source CLI. Review the quotas for your planned volume, since actions may be limited or throttled.

Questions people ask

Is there a free plan?

Yes. FREE costs nothing, billed monthly. Paid plans start at $10 a month.

What does the FREE plan include?

It includes 1 database, 50,000 AI tokens, 100 rows per table, 10 tables per database, 500 API requests, and 1 mock API with the listed endpoint, scenario, and request caps.

Which databases are supported?

Listed databases or destinations include MySQL, PostgreSQL, SQLite, MariaDB, MongoDB, and BigQuery.

Which platforms support the CLI?

The CLI has installation options for Linux, macOS, and Windows on amd64 and arm64.

What formats can be exported?

FakerForge offers SQL, JSON, CSV, and XML exports.

Does FakerForge provide API or MCP access?

Yes. FakerForge provides API access, and its MCP server is described as compatible with GitHub Copilot, Claude, Cursor, Windsurf, and other MCP-compatible clients.

What the maker publishes

AI data sharing
The privacy policy says schemas and messages may be shared with AI providers such as OpenAI to provide assistant features. source
AI generation
The service generates names, emails, addresses, timestamps, UUIDs and custom field types using AI-assisted mappings. source
API limit behavior
Rows are capped by plan limits, successful calls consume quota, and exceeding the API allowance returns HTTP 429. source
API mocking
Mock API endpoints can have scenarios for success states, errors, edge cases, and timeouts with configurable fake JSON payloads. source
API mocking
Mock endpoints support success, error, edge-case and timeout scenarios with configurable JSON payloads. source
Audience
The site identifies developers, QA engineers, data analysts, and product teams as users of FakerForge. source
CLI
The open-source CLI connects to private PostgreSQL or MySQL databases and pushes generated rows directly into tables. source
CLI credential handling
The CLI stores database credentials and API keys in the operating-system keychain and sends only schema shape to FakerForge. source
CLI credentials
The site says CLI database credentials stay in the operating system keychain and only schema shape is uploaded. source
CLI databases
The open-source CLI introspects local or private PostgreSQL and MySQL databases and can push generated data into their tables. source
CLI platforms
The CLI page says it has install options for Linux, macOS, and Windows on amd64 and arm64. source
CLI platforms
The CLI is offered through an install script, Homebrew, Go installation and binaries for Linux, macOS and Windows on amd64 and arm64. source
Database integrations
Officially listed database targets include MySQL 5.7/8, PostgreSQL 13–16, SQLite, MariaDB 10/11, MongoDB and BigQuery. source
Database support
The site lists MySQL, PostgreSQL, SQLite, MariaDB, MongoDB, and BigQuery among its supported databases or destinations. source
Documented quota limits
Large row targets can be capped, generation can fail after quota exhaustion, mock creation can stop at plan maximums and high request volume can be throttled or rejected. source
Export formats
FakerForge offers SQL, JSON, and CSV exports, and its documentation also lists XML as an output format. source
Export formats
Exports include SQL, JSON, CSV, XML, BSON-ready JSON, newline-delimited JSON and Parquet. source
Independent package extensibility
The separate Python project supports custom Faker providers, seeded generation, schema validation and optional NumPy, pandas and YAML extras. source
Independent package identity
A separate PyPI project also named FakerForge describes itself as a Python synthetic-data framework built on Faker; the page provides no link to fakerforge.com. source
Independent package licence
That separate PyPI project declares an MIT licence. source
Independent package maturity
The separate PyPI project is classified as Alpha and lists one maintainer. source
Independent package popularity
The separate GitHub repository shows one star and zero forks. source
Independent package release
PyPI records version 0.2.0 of the separate package as released on October 5, 2026. source
Independent package requirement
That separate PyPI project requires Python 3.10 or newer. source

Checked against the maker's pages on 15 Oct 2026.