--- library_name: transformers datasets: - radm/r1-multilingual-prefs-llama base_model: - deepseek-ai/DeepSeek-R1-Distill-Llama-8B --- # radm/DeepSeek-R1-Distill-Llama-8B-orpo Improved multilingual support using ORPO and LoRA based on dataset [radm/r1-multilingual-prefs-llama](https://huggingface.co/datasets/radm/r1-multilingual-prefs-llama)