This model is a fine-tuned model based on chatGLM2-6b with QLoRA. The only difference between this model and chatGLM2-6b should be the knowledge of "LoRA" and 'QLoRA' technique.