CohereLabs/global-mgsm
Global-MGSM Dataset Description Global-MGSM is a unified multilingual grade-school math benchmark built from the Multilingual Grade School Math (MGSM) benchmark and its extensions. Each language is provided as a separate config (subset). All language codes use ISO 639-1 two-letter codes. Languages (41 configs) Code Language Code Language Code Language ar Arabic bn Bengali ca Catalan cs Czech cy Welsh de German el Greek en English es… See the full description on the dataset page: https://huggingface.co/datasets/CohereLabs/global-mgsm.
| Author / organization | CohereLabs |
|---|---|
| Downloads | 880 (as reported by Hugging Face) |
| Likes | 1 |
| Created | 6 March 2026 |
| Last updated at source | 6 March 2026 |
Popularity, last 2 days
880 now (-25 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 16 Oct 2026 Description: Global-MGSM Dataset Description Global-MGSM is a unified multilingual grade-school math benchmark built from the Multilingual Grade School Math (MGSM) benchmark and its extensions. Each language is provided as a separate config (subset). Al → Global-MGSM Dataset Description Global-MGSM is a unified multilingual grade-school math benchmark built from the Multilingual Grade School Math (MGSM) benchmark and its extensions. Each language is provided as a separate config (subset). Al
Changes detected by Silytics between refreshes of the source.
More from CohereLabs
- CohereLabs/xP3xDataset
- CohereLabs/Global-MMLUDataset
- CohereLabs/aya_datasetDataset
- CohereLabs/aya_collectionDataset
- CohereLabs/Global-MMLU-LiteDataset
- CohereLabs/aya_collection_language_splitDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.