The research team of the Institute for Language and Speech Processing at the Athena Research Center presented Greece's second major AI language model, named Llama-Krikri, in the Monday, February 10 issue of Kathimerini, in an article by journalist Tasoula Karaiskaki.
Based on Meta’s open-source Llama 3.1, the new model succeeds Meltemi, offering significantly improved capabilities in understanding and generating Greek text. With enhanced memory and improved reasoning, it can handle longer conversations, support Ancient Greek, and provide more accurate responses.
Llama-Krikri is openly available, allowing researchers, businesses, and academic institutions to customize it according to their needs—without the high costs and restrictions of commercial models. Its development strengthens the digital survival of the Greek language, providing a technological solution that considers cultural identity and the needs of the local market.
The full interview with the ILSP/Athena team is available in
Kathimerini.
