Sign inGet started
Dataset · Hugging Face

large-traversaal/urdu-instruct

Urdu-Instruct Dataset Dataset Summary The Urdu-Instruct dataset is a high-quality multilingual synthetic corpus containing 51,686 training and 1,084 test examples. It was generated using GPT-4o under a modified self-instruction framework to improve instruction-following, reasoning, and bilingual understanding in Urdu. This dataset is part of the Alif-1.0-8B-Instruct project and was created by the Traversaal.AI Research Team. It supports seven essential… See the full description on the dataset page: https://huggingface.co/datasets/large-traversaal/urdu-instruct.

Author / organizationlarge-traversaal
License (as declared)cc-by-sa-4.0
Downloads375 (as reported by Hugging Face)
Likes1
Created11 March 2025
Last updated at source13 October 2025
Embed this entry
<iframe src="https://silytics.si/embed/item/30a0ab71-bf12-4215-b40e-d392485ba847" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

Popularity, last 3 days

375 now (+148 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.

Connected in the knowledge graph

Explore this entry in the graph

Recent changes

  • 17 Oct 2026 Description: Urdu-Instruct Dataset Dataset Summary The Urdu-Instruct dataset is a high-quality multilingual synthetic corpus containing 51,686 training and 1,084 test examples. It was generated using GPT-4o under a modified self-instruction framework to → Urdu-Instruct Dataset Dataset Summary The Urdu-Instruct dataset is a high-quality multilingual synthetic corpus containing 51,686 training and 1,084 test examples. It was generated using GPT-4o under a modified self-instruction framework to

Changes detected by Silytics between refreshes of the source.

More from large-traversaal

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.