Back to Gap Map Model components / Language-specific datasets

FLORES+

gated / Overall score: 4.4(strong)

FLORES+ is a multilingual machine translation benchmark of sentences sampled from Wikinews, Wikijunior and Wikivoyage and translated from English into more than 200 language varieties, with dev and devtest splits aligned across every language. It continues Meta's FLORES-200, and a separate blind test set stays with Meta. The Open Language Data Initiative maintains it and accepts community fixes and new languages.

Openness

3 high confidence

3.0

license: cc-by-sa-4.0
access: auto(Automatic Hugging Face gate: agree not to re-host where crawlers can reach it and to keep its contents out of training data when evaluating.)
dataset_card: present

The sentences carry a share-alike license and the gate approves anyone who accepts its terms. Those terms ask users not to re-host the files where crawlers can find them and to keep them out of training data, to protect the benchmark from contamination.

https://huggingface.co/api/datasets/openlanguagedata/flores_plus recorded 2026-09-24
gated: "auto"; gate fields: "I agree not to re-host FLORES+ in places where it could be picked up by web crawlers"; "If I evaluate using FLORES+, I will ensure that its contents are not in the training data".
https://huggingface.co/datasets/openlanguagedata/flores_plus recorded 2026-09-24
"FLORES+ is a multilingual machine translation benchmark released under CC BY-SA 4.0"; "It should not be used as training data."

Adoption

3 high confidence

3.0

Hugging Face downloads over the trailing month for the one FLORES+ repository. Copies bundled inside evaluation harnesses and earlier FLORES-200 downloads from Meta are not counted.

https://huggingface.co/api/datasets/openlanguagedata/flores_plus recorded 2026-09-24
12666 downloads in the trailing 30 days for openlanguagedata/flores_plus

Capability

5 medium confidence

5.0

FLORES+ is the reference translation benchmark for low-resource languages, with the same human-translated sentences in 230 varieties and documentation in the FLORES and NLLB papers. Meta evaluated NLLB-200 on it, and other benchmarks reuse its sentences, among them FLEURS for speech and Belebele for reading comprehension.

https://arxiv.org/abs/2205.12446 recorded 2026-09-24
FLEURS "is an n-way parallel speech dataset in 102 languages built on top of the machine translation FLoRes-101 benchmark".
https://arxiv.org/abs/2207.04672 recorded 2026-09-24
"we evaluated the performance of over 40,000 different translation directions using a human-translated benchmark, Flores-200".
https://huggingface.co/datasets/openlanguagedata/flores_plus recorded 2026-09-24
"For each language, the dataset has 997 sentences for the dev split and 1012 sentences for the devtest split"; "Currently 230 language varieties".
https://raw.githubusercontent.com/facebookresearch/belebele/main/README.md recorded 2026-09-24
"Each question has four multiple-choice answers and is linked to a short passage from the FLORES-200 dataset."

Verified 2026-09-24