# Complex task handling

Fused from multiple 12B-parameter large language models, equipped with text generation and dialogue capabilities

Kunoichi-DPO-v2-7B is a 7B-parameter large language model based on the Mistral architecture, optimized with Direct Preference Optimization (DPO) training, demonstrating outstanding performance in multiple benchmarks.