VLSP2025-LegalSML/qwen3-1.7b-legal-pretrain
TEXT GENERATIONConcurrency Cost:1Model Size:2BQuant:BF16Ctx Length:32kPublished:Jul 2, 2025Architecture:Transformer Warm

VLSP2025-LegalSML/qwen3-1.7b-legal-pretrain is a 1.7 billion parameter Qwen3-based language model developed by VLSP 2025 LegalSLM Task Organizers. It is continually pretrained on approximately 144,000 Vietnamese legal texts, including official documents and news articles, to specialize in Vietnamese legal text understanding and legal question answering. The model features a 32768 token context length and is optimized for legal domain-specific tasks in Vietnamese.

Loading preview...