NeMo Data Designer · AI Nav City

Core facts

Read by a script from the official API, not rewritten by a model.

Repository: https://github.com/NVIDIA-NeMo/DataDesigner
License: Apache-2.0
Languages: Python
Maintainer: NVIDIA-NeMo
Homepage: https://docs.nvidia.com/nemo/datadesigner
Verified: 2026-11-01

What it is

NeMo Data Designer generates synthetic datasets from scratch or using your own seed data. Data Designer provides the generation framework. Plugins extend the data sources, generation capabilities, and processing steps available in Data Designer workflows. NeMo Data Designer is licensed under Apache License 2.0.

Main features

Data Designer generates text, structured data, and images using samplers, LLMs, image models, or existing seed datasets.
Data Designer builds multimodal workflows with image, audio, and video context.
Data Designer connects generation to external tools through local or remote MCP servers and captures interaction traces.
Data Designer controls relationships between fields with dependency-aware generation.
Data Designer validates and scores outputs with Python, SQL, custom validators, and LLM judges.
Data Designer previews, resumes, and monitors generation, from small experiments to large runs.

Getting started

Data Designer can also be installed from source.
The optional security constraints file applies maintained minimum dependency versions for known vulnerabilities.
The project lists default model providers to start with.
After installing Data Designer, compatible packages can be browsed and installed from the default NVIDIA catalog.
The NVIDIA catalog is configured by default.

From the README · bash

pip install data-designer

From the README · bash

git clone https://github.com/NVIDIA-NeMo/DataDesigner.git
cd DataDesigner
make install

From the README · bash

export NVIDIA_API_KEY="your-api-key-here"

export OPENAI_API_KEY="your-openai-api-key-here"

export OPENROUTER_API_KEY="your-openrouter-api-key-here"

From the README · python

import data_designer.config as dd
from data_designer.interface import DataDesigner

# Initialize with default settings
data_designer = DataDesigner()
config_builder = dd.DataDesignerConfigBuilder()

# Add a product category
config_builder.add_column(
    dd.SamplerColumnConfig(
        name="product_category",
        sampler_type=dd.SamplerType.CATEGORY,
        params=dd.CategorySamplerParams(
            values=["Electronics", "Clothing", "Home & Kitchen", "Books"],
        ),
    )
)

# Generate personalized customer reviews
config_builder.add_column(
    dd.LLMTextColumnConfig(
        name="review",
        model_alias="nvidia-text",
        prompt="Write a brief product review for a {{ product_category }} item you recently purchased.",
    )
)

# Preview your dataset
preview = data_designer.preview(config_builder=config_builder)
preview.display_sample_record()

From the README · bash

data-designer plugin list
data-designer plugin info github
data-designer plugin install github
data-designer plugin installed

Who it is for

The DataDesignerPlugins repository contains optional NVIDIA-maintained packages that add plugin capabilities. Plugin packages add specialized capabilities to an existing workflow and can also provide ready-made workflows with their own CLI. Plugin packages are installed in the same Python environment as Data Designer.

The extensions of plugin packages are discovered automatically after installation. Data Designer has a skill for coding agents. The skill is tested with Claude Code and Codex.

Data Designer collects telemetry to help improve the library. NeMo Data Designer includes an optional function to share anonymous telemetry data with NVIDIA for product improvement. The telemetry data collected is limited to names of models used and token counts (input and output).

Telemetry does not collect user or device information. Users may opt out of telemetry collection at any time. Opting out applies only to data collection by the NeMo Data Designer library itself.

NeMo Data Designer can be configured to use various inference endpoints. NVIDIA Build is intended for evaluation and testing purposes only and may not be used in production environments.

Related tools listed on this site.