Dicta-il/DictaLM-3.0-1.7B-Instruct is a 1.7 billion parameter instruction-tuned large language model developed by Dicta, initialized from Qwen3-1.7B-Base. This model is specifically optimized for Hebrew language tasks, setting a new state-of-the-art for its weight class in Hebrew as both a base and chat model. It supports a 40960-token context length and is designed for general-purpose conversational AI and applications requiring strong Hebrew language understanding and generation.
No reviews yet. Be the first to review!