KaraKaraWitch/GoldDiamondGold-L33-70b

KaraKaraWitch/GoldDiamondGold-L33-70b is a 70 billion parameter merged language model, built upon the deepcogito/cogito-v2-preview-llama-70B base using the SCE merge method. This model integrates components from numerous Llama-family models, focusing on enhanced intelligence, nuanced prose style, and improved roleplaying capabilities. It is designed to be responsive to well-crafted system prompts and aims for a balanced output, despite exhibiting a slight 'horny streak' and 'Claudisms'.