enstazao/Qalb-1.0-8B-Instruct
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Jan 14, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

Qalb-1.0-8B-Instruct is an 8 billion parameter Urdu language model developed by enstazao, built upon the Llama-3.1-8B architecture. It has undergone continued pre-training on a 1.97 billion token Urdu corpus and supervised fine-tuning for instruction following. This model excels in deep Urdu understanding, reasoning, and culturally accurate responses, outperforming previous state-of-the-art models in Urdu language tasks while retaining strong English capabilities.

Loading preview...