| Scenario | LoRA | QLoRA | PEFT | DeepSpeed | ZeRO | DORA |
| Adapting pre-trained LLMs to specific tasks or domains | 是 | 是 | 是 | 是 | 是 | 是 |
| Fine-tuning for NLP tasks such as text classification, named entity recognition, and machine translation | 是 | 是 | 是 | 是 | 是 | 是 |
| Fine-tuning for QA tasks | 是 | 是 | 是 | 是 | 是 | 是 |
| Fine-tuning for generating human-like responses in chatbots | 是 | 是 | 是 | 是 | 是 | 是 |
| Fine-tuning for generating music, art, or other forms of creativity | 是 | 是 | 是 | 是 | 是 | 是 |
| Reducing computational and financial costs | 是 | 是 | 否 | 是 | 是 | 否 |
| Reducing memory usage | 否 | 是 | 否 | 是 | 是 | 是 |
| Using fewer parameters for efficient finetuning | 否 | 是 | 是 | 否 | 否 | 是 |
| Memory-efficient form of data parallelism that gives access to the aggregate GPU memory of all the GPU devices available | 否 | 否 | 否 | 是 | 是 | 是 |
