DATAMIMIC CE

DATAMIMIC CE is a pip-installable Python developer tool for model-driven generation, SQL-selected source subsets and explicit field transformations. Its Python API, CLI and optional MCP adapter support local, CI and agent-assisted workflows. CE has its own execution core, separate from EE.

Sources reviewed

When should you consider it?

Consider it for customer-controlled XML models, reusable fixtures and explicit business rules in local development or CI. A downloaded Enterprise project can run without rewriting its model when all functions and dependencies are supported by CE.

Editorial fit assessment based on the sources below.

What are the limits?

PII field selection and pseudonymization models are manual. The EE ML engine, central Platform governance and browser IDE belong to the Enterprise product. Seeded replay evidence is scoped to the documented CE runtime matrix and outputs; a seed alone does not guarantee every exporter or environment.

A developer runtime with a distinct product scope

CE exposes local data generation and model tooling through Python, CLI and an optional agent adapter. Teams integrate it into their own development environment and automation. Central project governance, Platform IDE integration and managed enterprise task execution are assessed in the separate Enterprise Platform profile.

Reference

Documented SQL subsetting and transformation

The published CE 4.3.0 payment example selects payment IDs 1–3 at the SQLite source using WHERE and ORDER BY, then writes explicitly transformed records to a separate target table. The article reports that source row 4 is excluded. This documents model-defined source-row subsetting; it does not demonstrate CE automatically discovering and closing a multi-table foreign-key graph. The Platform planning workflow is evaluated separately.

Reference

Shared models, separate execution engines

CE can execute exported customer models within its supported DSL scope, subject to compatible versions, configuration and resources, and available dependencies, inputs and target services. Projects using EE-only Kafka, ML or advanced nodes need EE or changes to those functions. CE supports Python multiprocessing and optional Ray; its engine and performance profile differ from EE. Platform governance and execution provenance remain separate services.

Reference

Enterprise Platform and CE have separate profiles.

The Enterprise product combines an EE execution core with Platform services. CE is a separate installable developer package. A capability or proof from one runtime does not automatically establish it for the other.

DATAMIMIC Enterprise Platform

Which capabilities are documented?

Synthetic generationLocal CE models and API

De-identification / maskingManual model-authored transforms

Subsetting / subset planningSQL-defined source-row subsets

Data virtualizationNot verified

Seeded replay, scopedScoped CE runtime matrix

Documented means a cited vendor source describes this scoped capability. Not verified means the reviewed evidence cannot establish it. Neither is a hands-on test result.

How is it deployed and licensed?

Install with pip install datamimic-ce and run locally through Python or the datamimic CLI, including CI. The optional adapter is installed with pip install "datamimic-ce[mcp]". This local MCP adapter is distinct from the Platform project-scoped MCP service.

MIT-licensed Community Edition according to the project repository. The commercial Enterprise Platform and its separate EE core have their own licensing and deployment.

Sources and scope

rapiddweller/datamimic repository
CE package installation, Python/CLI APIs, optional local MCP adapter, MIT license, manual pseudonymization and the explicit CE replay test boundary.
Source checked: 2026-10-01
Upgrade your models from DATAMIMIC 3.5 to 4.0
Exact replay boundary; requires same engine version, complete deterministic inputs, explicit seed, execution topology, deterministic serialization and ordering; ML is excluded; date behavior depends on the seeded reference clock.
Source checked: 2026-10-01
Date and Time Generation
Relative date windows use a runtime clock anchor; seeded execution uses deterministic runtime clock, while unseeded execution reads live clock.
Source checked: 2026-10-01
DATAMIMIC vs Delphix: source selection and transformation example
Vendor-published CE 4.3.0 example: SQL range selection, explicit field transformations and target writing; edition-scoped four-path explanation. The reported execution was not independently repeated in this comparison.
Source checked: 2026-10-01
Database and other source selection contracts
SQL selectors and source read semantics; distinguishes Platform-owned environment configuration from model execution.
Source checked: 2026-10-01
Iterate source traversal
Source records and parent context for explicit child transformations and target generation.
Source checked: 2026-10-01

What should you verify in a proof of concept?

Your database version, schema constraints and target formats.
The exact product, edition, deployment and license entitlements.
Your business assertions and the meaning of repeatability for your output.
A failed run, cleanup and a repeat run on controlled inputs.