👋🤗🤗👋 Join our WeChat.
中文 | English
LLamaTuner is an efficient, flexible and full-featured toolkit for fine-tuning LLM (Llama3, Phi3, Qwen, Mistral, ...)
Efficient
- Support LLM, VLM pre-training / fine-tuning on almost all GPUs. LLamaTuner is capable of fine-tuning 7B LLM on a single 8GB GPU, as well as multi-node fine-tuning of models exceeding 70B.
- Automatically dispatch high-performance operators such as FlashAttention and Triton kernels to increase training throughput.
- Compatible with DeepSpeed 🚀, easily utilizing a variety of ZeRO optimization techniques.
Flexible
- Support various LLMs (Llama 3, Mixtral, Llama 2, ChatGLM, Qwen, Baichuan, ...).
- Support VLM (LLaVA).
- Well-designed data pipeline, accommodating datasets in any format, including but not limited to open-source and custom formats.
- Support various training algorithms (QLoRA, LoRA, full-parameter fune-tune), allowing users to choose the most suitable solution for their requirements.
Full-featured
- Support continuous pre-training, instruction fine-tuning, and agent fine-tuning.
- Support chatting with large models with pre-defined templates.
- Easy and Efficient Fine-tuning LLMs --- 简单高效的大语言模型训练/部署
| Model | Model size | Default module | Template |
|---|---|---|---|
| Baichuan | 7B/13B | W_pack | baichuan |
| Baichuan2 | 7B/13B | W_pack | baichuan2 |
| BLOOM | 560M/1.1B/1.7B/3B/7.1B/176B | query_key_value | - |
| BLOOMZ | 560M/1.1B/1.7B/3B/7.1B/176B | query_key_value | - |
| ChatGLM3 | 6B | query_key_value | chatglm3 |
| Command-R | 35B/104B | q_proj,v_proj | cohere |
| DeepSeek (MoE) | 7B/16B/67B/236B | q_proj,v_proj | deepseek |
| Falcon | 7B/11B/40B/180B | query_key_value | falcon |
| Gemma/CodeGemma | 2B/7B | q_proj,v_proj | gemma |
| InternLM2 | 7B/20B | wqkv | intern2 |
| LLaMA | 7B/13B/33B/65B | q_proj,v_proj | - |
| LLaMA-2 | 7B/13B/70B | q_proj,v_proj | llama2 |
| LLaMA-3 | 8B/70B | q_proj,v_proj | llama3 |
| LLaVA-1.5 | 7B/13B | q_proj,v_proj | vicuna |
| Mistral/Mixtral | 7B/8x7B/8x22B | q_proj,v_proj | mistral |
| OLMo | 1B/7B | q_proj,v_proj | - |
| PaliGemma | 3B | q_proj,v_proj | gemma |
| Phi-1.5/2 | 1.3B/2.7B | q_proj,v_proj | - |
| Phi-3 | 3.8B | qkv_proj | phi |
| Qwen | 1.8B/7B/14B/72B | c_attn | qwen |
| Qwen1.5 (Code/MoE) | 0.5B/1.8B/4B/7B/14B/32B/72B/110B | q_proj,v_proj | qwen |
| StarCoder2 | 3B/7B/15B | q_proj,v_proj | - |
| XVERSE | 7B/13B/65B | q_proj,v_proj | xverse |
| Yi (1/1.5) | 6B/9B/34B | q_proj,v_proj | yi |
| Yi-VL | 6B/34B | q_proj,v_proj | yi_vl |
| Yuan | 2B/51B/102B | q_proj,v_proj | yuan |
| Approach | Full-tuning | Freeze-tuning | LoRA | QLoRA |
|---|---|---|---|---|
| Pre-Training | ✅ | ✅ | ✅ | ✅ |
| Supervised Fine-Tuning | ✅ | ✅ | ✅ | ✅ |
| Reward Modeling | ✅ | ✅ | ✅ | ✅ |
| PPO Training | ✅ | ✅ | ✅ | ✅ |
| DPO Training | ✅ | ✅ | ✅ | ✅ |
| KTO Training | ✅ | ✅ | ✅ | ✅ |
| ORPO Training | ✅ | ✅ | ✅ | ✅ |
As of now, we support the following datasets, most of which are all available in the Hugging Face datasets library.
Supervised fine-tuning dataset
- Stanford Alpaca
- Stanford Alpaca (Chinese)
- Hello-SimpleAI/HC3
- BELLE 2M (zh)
- BELLE 1M (zh)
- BELLE 0.5M (zh)
- BELLE Dialogue 0.4M (zh)
- BELLE School Math 0.25M (zh)
- BELLE Multiturn Chat 0.8M (zh)
- databricks-dolly-15k
- mosaicml/dolly_hhrlhf
- GPT-4 Generated Data
- Alpaca CoT
- UltraChat
- OpenAssistant/oasst1
- ShareGPT_Vicuna_unfiltered
- BIAI/OL-CC
- timdettmers/openassistant-guanaco
- Evol-Instruct
- OpenOrca
- Platypus
- OpenHermes
Preference datasets
Please refer to data/README.md to learn how to use these datasets. If you want to explore more datasets, please refer to the awesome-instruction-datasets. Some datasets require confirmation before using them, so we recommend logging in with your Hugging Face account using these commands.
pip install --upgrade huggingface_hub
huggingface-cli loginWe provide a number of data preprocessing tools in the data folder. These tools are intended to be a starting point for further research and development.
- data_utils.py : Data preprocessing and formatting
- sft_dataset.py : Supervised fine-tuning dataset class and collator
- conv_dataset.py : Conversation dataset class and collator
We provide a number of models in the Hugging Face model hub. These models are trained with QLoRA and can be used for inference and finetuning. We provide the following models:
| Base Model | Adapter | Instruct Datasets | Train Script | Log | Model on Huggingface |
|---|---|---|---|---|---|
| llama-7b | FullFinetune | - | - | - | |
| llama-7b | QLoRA | openassistant-guanaco | finetune_lamma7b | wandb log | GaussianTech/llama-7b-sft |
| llama-7b | QLoRA | OL-CC | finetune_lamma7b | ||
| baichuan7b | QLoRA | openassistant-guanaco | finetune_baichuan7b | wandb log | GaussianTech/baichuan-7b-sft |
| baichuan7b | QLoRA | OL-CC | finetune_baichuan7b | wandb log | - |
| Mandatory | Minimum | Recommend |
|---|---|---|
| python | 3.8 | 3.10 |
| torch | 1.13.1 | 2.2.0 |
| transformers | 4.37.2 | 4.41.0 |
| datasets | 2.14.3 | 2.19.1 |
| accelerate | 0.27.2 | 0.30.1 |
| peft | 0.9.0 | 0.11.1 |
| trl | 0.8.2 | 0.8.6 |
| Optional | Minimum | Recommend |
|---|---|---|
| CUDA | 11.6 | 12.2 |
| deepspeed | 0.10.0 | 0.14.0 |
| bitsandbytes | 0.39.0 | 0.43.1 |
| vllm | 0.4.0 | 0.4.2 |
| flash-attn | 2.3.0 | 2.5.8 |
* estimated
| Method | Bits | 7B | 13B | 30B | 70B | 110B | 8x7B | 8x22B |
|---|---|---|---|---|---|---|---|---|
| Full | AMP | 120GB | 240GB | 600GB | 1200GB | 2000GB | 900GB | 2400GB |
| Full | 16 | 60GB | 120GB | 300GB | 600GB | 900GB | 400GB | 1200GB |
| Freeze | 16 | 20GB | 40GB | 80GB | 200GB | 360GB | 160GB | 400GB |
| LoRA/GaLore/BAdam | 16 | 16GB | 32GB | 64GB | 160GB | 240GB | 120GB | 320GB |
| QLoRA | 8 | 10GB | 20GB | 40GB | 80GB | 140GB | 60GB | 160GB |
| QLoRA | 4 | 6GB | 12GB | 24GB | 48GB | 72GB | 30GB | 96GB |
| QLoRA | 2 | 4GB | 8GB | 16GB | 24GB | 48GB | 18GB | 48GB |
Clone this repository and navigate to the Efficient-Tuning-LLMs folder
git clone https://github.com/jianzhnie/LLamaTuner.git
cd LLamaTuner| main function | Useage | Scripts |
|---|---|---|
| train.py | Full finetune LLMs on SFT datasets | full_finetune |
| train_lora.py | Finetune LLMs by using Lora (Low-Rank Adaptation of Large Language Models finetune) | lora_finetune |
| train_qlora.py | Finetune LLMs by using QLora (QLoRA: Efficient Finetuning of Quantized LLMs) | qlora_finetune |
The train_qlora.py code is a starting point for finetuning and inference on various datasets.
Basic command for finetuning a baseline model on the Alpaca dataset:
python train_qlora.py --model_name_or_path <path_or_name>For models larger than 13B, we recommend adjusting the learning rate:
python train_qlora.py –learning_rate 0.0001 --model_name_or_path <path_or_name>To find more scripts for finetuning and inference, please refer to the scripts folder.
Here a list of known issues and bugs. If your issue is not reported here, please open a new issue and describe the problem.
- 4-bit inference is slow. Currently, our 4-bit inference implementation is not yet integrated with the 4-bit matrix multiplication
- Resuming a LoRA training run with the Trainer currently runs on an error
- Currently, using
bnb_4bit_compute_type='fp16'can lead to instabilities. For 7B LLaMA, only 80% of finetuning runs complete without error. We have solutions, but they are not integrated yet into bitsandbytes. - Make sure that
tokenizer.bos_token_id = 1to avoid generation issues.
LLamaTuner is released under the Apache 2.0 license.
We thank the Huggingface team, in particular Younes Belkada, for their support integrating QLoRA with PEFT and transformers libraries.
We appreciate the work by many open-source contributors, especially:
- LLaMa
- Vicuna
- xTuring
- Alpaca-LoRA
- Stanford Alpaca
- LLaMA-Factory
- Hugging Face
- Peft
- axolotl
- deepspeed
- Unsloth
- qlora
- bitsandbytes
- https://github.com/QwenLM
- https://github.com/InternLM
- https://github.com/ymcui/Chinese-LLaMA-Alpaca-3
- https://github.com/ymcui/Chinese-Mixtral
- https://github.com/SmartFlowAI/EmoLLM
- https://github.com/yangjianxin1/Firefly
- https://github.com/LiuHC0428/LAW-GPT
Please cite the repo if you use the data or code in this repo.
@misc{Chinese-Guanaco,
author = {jianzhnie},
title = {LLamaTuner: Easy and Efficient Fine-tuning LLMs},
year = {2023},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/jianzhnie/LLamaTuner}},
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent
