This repository provides tools and implementations for fine-tuning ChatGLM models, including support for GLM-4 variants. It supports multiple fine-tuning methods such as LoRA, QLoRA, AdaLoRA, IA3, and P-tuning v2, enabling efficient training of large language models with reduced computational overhead. The framework integrates with transformers, DeepSpeed, and bitsandbytes for distributed training and quantization support.