FLORES-200 Dataset
by Meta AI · open-source · Last verified 2026-03-17
FLORES-200 is Meta's few-shot translation evaluation benchmark spanning 200 languages, including many low-resource and endangered ones. Each language contains 1,012 parallel sentences translated from English Wikipedia, covering both devtest and test splits for systematic MT evaluation at scale.
https://huggingface.co/datasets/facebook/flores ↗B+
B+—Good
Adoption: AQuality: A+Freshness: B+Citations: AEngagement: F
Specifications
- License
- CC-BY-SA-4.0
- Pricing
- open-source
- Capabilities
- translation-evaluation, multilingual-benchmark, low-resource-evaluation
- Integrations
- huggingface-datasets
- Use Cases
- machine-translation, model-evaluation, low-resource-nlp
- API Available
- No
- Tags
- evaluation, machine-translation, 200-languages, low-resource, meta
- Added
- 2026-03-17
- Completeness
- 100%
Index Score
73Adoption
82
Quality
91
Freshness
75
Citations
88
Engagement
0
Put AI to work for your business
Deploy this dataset alongside autonomous AaaS agents that handle tasks end-to-end — no babysitting required.