KaraKaraWitch/Matsutei-Qwen2.5-72b

KaraKaraWitch/Matsutei-Qwen2.5-72b is a 72.7 billion parameter language model based on the Qwen2.5 architecture, created by KaraKaraWitch through a TIES merge of several pre-trained models. This model is designed to address specific issues related to world book lore information injection, aiming to reduce confusion when integrating external context. With a substantial 131,072 token context length, it is optimized for handling extensive narrative and contextual data. Its primary strength lies in managing complex lore and preventing initial confusion during content generation.