Skip to content

Dataset · nvidia

Synthetic personas of German residents by NVIDIA

nvidia/Nemotron-Personas-Germany

1 to 10 million descriptions of virtual German residents generated based on real demographics.

What's inside

The dataset from NVIDIA contains from 1 to 10 million text descriptions of synthetic personas in German. The profiles are generated using a Compound-AI approach based on real demographic and geographic population distributions of Germany. The data is distributed under the open CC BY 4.0 license.

What you can do with it

  • Generation of realistic synthetic dialogues and user scenarios in German
  • Simulation of behavior and surveys of a diverse audience when testing LLM agents
  • Fine-tuning of language models for response personalization and roleplay scenarios
Tasks
Text generation
Languages
German
Size
1M–10M rows
Downloads this month
2.8K
Likes
53

How to load it

from datasets import load_dataset

ds = load_dataset("nvidia/Nemotron-Personas-Germany")