v000000/Qwen2.5-Lumen-14B

Qwen2.5-Lumen-14B is a 14.8 billion parameter language model based on the Qwen2.5 architecture, fine-tuned using direct preference optimization (DPO) for approximately three epochs. This model specializes in prompt adherence, story writing, and roleplay, leveraging a merge of multiple DPO checkpoints and SLERP variations. It supports a full context length of 131,072 tokens, making it particularly suitable for generating long-form narrative content and engaging in character-based interactions.