LumiOpen/Llama-Poro-2-70B-Instruct

LumiOpen/Llama-Poro-2-70B-Instruct is a 70.55 billion parameter instruction-following chatbot model developed by AMD Silo AI, TurkuNLP, and HPLT. Based on Llama 3.1 70B, it was created through continued pretraining, supervised fine-tuning (SFT), and Direct Preference Optimization (DPO) on a mix of English and Finnish instruction data. This model excels at conversational AI and instruction following, demonstrating substantial improvements in Finnish capabilities while maintaining strong English performance.