robinsmits/Qwen1.5-7B-Dutch-Chat

The robinsmits/Qwen1.5-7B-Dutch-Chat is a 7.7 billion parameter DPO-aligned language model based on the Qwen1.5 architecture, fine-tuned specifically for the Dutch language. It leverages the Qwen/Qwen1.5-7B-Chat as its base model and was trained on the Dutch ultra_feedback_dutch_cleaned dataset. This model demonstrates performance in Dutch natural language understanding and generation tasks comparable to GPT-3.5, making it suitable for Dutch-centric conversational AI applications.