Onomaverse Names Datasets

Open data on 38,069 names across 106 countries and 94 languages: name frequencies, multilingual transliterations, gender inference, a name-equivalence graph, and name-day calendars.

🔗 Source & full documentation: onomaverse.com/datasets 📄 License: CC BY 4.0 · 🆔 DOI: 10.5281/zenodo.20797446

Names data from Onomaverse, licensed CC BY 4.0.

Subsets

Each subset is a separate config — load any of them by name. Per-subset documentation lives on the website.

Quick start

from datasets import load_dataset

# Pick any subset by its config name:
ds = load_dataset("onomaverse/names", "name-transliterations", split="train")
print(ds[0])

gnf = load_dataset("onomaverse/names", "given-name-frequency", split="train")

Every subset also ships as CSV / JSON Lines / Parquet in this repo, so you can grab files directly:

import pandas as pd
df = pd.read_parquet("hf://datasets/onomaverse/names/given-name-frequency.parquet")

Attribution (required)

Names data from Onomaverse, licensed CC BY 4.0. A link back to onomaverse.com/datasets satisfies the attribution requirement.

This dataset contains no descriptive prose (meanings, etymology, biographies). For that — for every name, in 94 languages — visit onomaverse.com.