[](https://github.com/hiyouga/LLaMA-Factory/stargazers)
[](LICENSE)
[](https://github.com/hiyouga/LLaMA-Factory/commits/main)
[](https://pypi.org/project/llmtuner/)
[](https://pypi.org/project/llmtuner/)
[](#projects-using-llama-factory)
[](https://github.com/hiyouga/LLaMA-Factory/pulls)
[](https://discord.gg/rKfvV9r9FK)
[](https://twitter.com/llamafactory_ai)
[](https://huggingface.co/spaces/hiyouga/LLaMA-Board)
[](https://modelscope.cn/studios/hiyouga/LLaMA-Board)
[](https://colab.research.google.com/drive/1eRTPn37ltBbYsISy9Aw2NuI2Aq5CQrD9?usp=sharing)
π Join our [WeChat](assets/wechat.jpg).
\[ English | [δΈζ](README_zh.md) \]
**Fine-tuning a large language model can be easy as...**
https://github.com/hiyouga/LLaMA-Factory/assets/16256802/9840a653-7e9c-41c8-ae89-7ace5698baf6
Choose your path:
- **Colab**: https://colab.research.google.com/drive/1eRTPn37ltBbYsISy9Aw2NuI2Aq5CQrD9?usp=sharing
- **Local machine**: Please refer to [usage](#getting-started)
## Table of Contents
- [Features](#features)
- [Benchmark](#benchmark)
- [Changelog](#changelog)
- [Supported Models](#supported-models)
- [Supported Training Approaches](#supported-training-approaches)
- [Provided Datasets](#provided-datasets)
- [Requirement](#requirement)
- [Getting Started](#getting-started)
- [Projects using LLaMA Factory](#projects-using-llama-factory)
- [License](#license)
- [Citation](#citation)
- [Acknowledgement](#acknowledgement)
## Features
- **Various models**: LLaMA, LLaVA, Mistral, Mixtral-MoE, Qwen, Yi, Gemma, Baichuan, ChatGLM, Phi, etc.
- **Integrated methods**: (Continuous) pre-training, (multimodal) supervised fine-tuning, reward modeling, PPO, DPO and ORPO.
- **Scalable resources**: 32-bit full-tuning, 16-bit freeze-tuning, 16-bit LoRA and 2/4/8-bit QLoRA via AQLM/AWQ/GPTQ/LLM.int8.
- **Advanced algorithms**: GaLore, BAdam, DoRA, LongLoRA, LLaMA Pro, Mixture-of-Depths, LoRA+, LoftQ and Agent tuning.
- **Practical tricks**: FlashAttention-2, Unsloth, RoPE scaling, NEFTune and rsLoRA.
- **Experiment monitors**: LlamaBoard, TensorBoard, Wandb, MLflow, etc.
- **Faster inference**: OpenAI-style API, Gradio UI and CLI with vLLM worker.
## Benchmark
Compared to ChatGLM's [P-Tuning](https://github.com/THUDM/ChatGLM2-6B/tree/main/ptuning), LLaMA Factory's LoRA tuning offers up to **3.7 times faster** training speed with a better Rouge score on the advertising text generation task. By leveraging 4-bit quantization technique, LLaMA Factory's QLoRA further improves the efficiency regarding the GPU memory.
<details><summary>Definitions</summary>
- **Training Speed**: the number of training samples processed per second during the training. (bs=4, cutoff_len=1024)
- **Rouge Score**: Rouge-2 score on the development set of the [advertising text generation](https://aclanthology.org/D19-1321.pdf) task. (bs=4, cutoff_len=1024)
- **GPU Memory**: Peak GPU memory usage in 4-bit quantized training. (bs=1, cutoff_len=1024)
- We adopt `pre_seq_len=128` for ChatGLM's P-Tuning and `lora_rank=32` for LLaMA Factory's LoRA tuning.
</details>
## Changelog
[24/04/26] We supported fine-tuning the **LLaVA-1.5** multimodal LLMs. See `examples/lora_single_gpu/sft_mllm.sh` for usage.
[24/04/22] We provided a **[Colab notebook](https://colab.research.google.com/drive/1eRTPn37ltBbYsISy9Aw2NuI2Aq5CQrD9?usp=sharing)** for fine-tuning the Llama-3 model on a free T4 GPU. Two Llama-3-derived models fine-tuned using LLaMA Factory are available at Hugging Face, check [Llama3-8B-Chinese-Chat](https://huggingface.co/shenzhi-wang/Llama3-8B-Chinese-Chat) and [Llama3-Chinese](https://huggingface.co/zhichen/Llama3-Chinese) for details.
[24/04/21] We supported **[Mixture-of-Depths](https://arxiv.org/abs/2404.02258)** according to [AstraMindAI's implementation](https://github.com/astramind-ai/Mixture-of-depths). See `examples/extras/mod` for usage.
[24/04/16] We supported **[BAdam](https://arxiv.org/abs/2404.02827)**. See `examples/extras/badam` for usage.
[24/04/16] We supported **[unsloth](https://github.com/unslothai/unsloth)**'s long-sequence training (Llama-2-7B-56k within 24GB). It achieves **117%** speed and **50%** memory compared with FlashAttention-2, more benchmarks can be found in [this page](https://github.com/hiyouga/LLaMA-Factory/wiki/Performance-comparison).
<details><summary>Full Changelog</summary>
[24/03/31] We supported **[ORPO](https://arxiv.org/abs/2403.07691)**. See `examples/lora_single_gpu` for usage.
[24/03/21] Our paper "[LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://arxiv.org/abs/2403.13372)" is available at arXiv!
[24/03/20] We supported **FSDP+QLoRA** that fine-tunes a 70B model on 2x24GB GPUs. See `examples/extras/fsdp_qlora` for usage.
[24/03/13] We supported **[LoRA+](https://arxiv.org/abs/2402.12354)**. See `examples/extras/loraplus` for usage.
[24/03/07] We supported gradient low-rank projection (**[GaLore](https://arxiv.org/abs/2403.03507)**) algorithm. See `examples/extras/galore` for usage.
[24/03/07] We integrated **[vLLM](https://github.com/vllm-project/vllm)** for faster and concurrent inference. Try `--infer_backend vllm` to enjoy **270%** inference speed. (LoRA is not yet supported, merge it first.)
[24/02/28] We supported weight-decomposed LoRA (**[DoRA](https://arxiv.org/abs/2402.09353)**). Try `--use_dora` to activate DoRA training.
[24/02/15] We supported **block expansion** proposed by [LLaMA Pro](https://github.com/TencentARC/LLaMA-Pro). See `examples/extras/llama_pro` for usage.
[24/02/05] Qwen1.5 (Qwen2 beta version) series models are supported in LLaMA-Factory. Check this [blog post](https://qwenlm.github.io/blog/qwen1.5/) for details.
[24/01/18] We supported **agent tuning** for most models, equipping model with tool using abilities by fine-tuning with `--dataset glaive_toolcall`.
[23/12/23] We supported **[unsloth](https://github.com/unslothai/unsloth)**'s implementation to boost LoRA tuning for the LLaMA, Mistral and Yi models. Try `--use_unsloth` argument to activate unsloth patch. It achieves **170%** speed in our benchmark, check [this page](https://github.com/hiyouga/LLaMA-Factory/wiki/Performance-comparison) for details.
[23/12/12] We supported fine-tuning the latest MoE model **[Mixtral 8x7B](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1)** in our framework. See hardware requirement [here](#hardware-requirement).
[23/12/01] We supported downloading pre-trained models and datasets from the **[ModelScope Hub](https://modelscope.cn/models)** for Chinese mainland users. See [this tutorial](#use-modelscope-hub-optional) for usage.
[23/10/21] We supported **[NEFTune](https://arxiv.org/abs/2310.05914)** trick for fine-tuning. Try `--neftune_noise_alpha` argument to activate NEFTune, e.g., `--neftune_noise_alpha 5`.
[23/09/27] We supported **$S^2$-Attn** proposed by [LongLoRA](https://github.com/dvlab-research/LongLoRA) for the LLaMA models. Try `--shift_attn` argument to enable shift short attention.
[23/09/23] We integrated MMLU, C-Eval and CMMLU benchmarks in this repo. See [this example](#evaluation) to evaluate your models.
[23/09/10] We supported **[FlashAttention-2](https://github.com/Dao-AILab/flash-attention)**. Try `--flash_attn fa2` argument to enable FlashAttention-2 if you are using RTX4090, A100 or H100 GPUs.
[23/08/12] We supported **RoPE scaling** to extend the context length of the LLaMA models. Try `--rope_scaling linear` argument in training and `--rope_scaling dynamic` argument at inference to extrapolate the position embeddings.
[23/08/11] We supported **[DPO training](https://arxiv.org/abs/2305.18290)** for instruction-tuned models. See [this example](#dpo-training) to train your models.
[23/07/31] We supported **dataset streaming**. Try `--streaming` and `--max_steps 10000` arguments to load your dataset in streaming mode.
[23/07/29] We released two instruction-tuned 13B models at Hugging Face. See these Hugging Face Repos ([LLaMA-2](https://huggingface.co/hiyouga/Llama-2-Chinese-13b-chat) / [Baichuan](https://huggingface.co/hiyouga/Baichuan-13B-sft)) for details.
[23/07/18] We developed an **all-in-one Web UI** for training, evaluation and inference. Try `train_web.py` to fine-tune models in your Web browser. Thank [@KanadeSiina](https://github.com/KanadeSiina) and [@codemayq](https://github.com/codemayq) for their efforts in the development.
[23/07/09] We released **[FastEdit](https://github.com/hiyouga/FastEdit)** β‘π©Ή, an easy-to-use package for editing the factual knowledge of large language models efficiently. Please follow [FastEdit](https://github.com/hiyouga/FastEdit) if you are interested.
[23/06/29] We provided a **reproducible example** of training a chat model using instruction-following datasets, see [Baichuan-7B-sft](https://huggingface.co/hiyouga/Baichuan-7B-sft) for details.
[23/06/22] We aligned the [demo API](src/api_demo.py) with the [OpenAI's](https://platform.openai.com/docs/api-reference/chat) format where you can insert the fine-tuned model in **arbitrary ChatGPT-based applications**.
[23/06/03] We supported quantized training and inference (aka **[QLoRA](https://github.com/artidoro/qlora)**). Try `--quantization_bit 4/8` argument to work with quantized models.
</details>
## Supported Models
| Model | Model size | Default module | Template |
| -------------------------------------------------------- | -------------------------------- | ----------------- | --------- |
| [Baichuan2](https://huggingface.co/baichuan-inc) | 7B/13B | W_pack | baichuan2 |
| [BLOOM](https://huggingface.co/bigscience) | 560M/1.1B/1.7B/3B/7.1B/176B | query_key_value | - |
| [BLOOMZ](https://huggingface.co/bigscience) | 560M/1.1B/1.7B/3B/7.1B/176B | query_key_value | - |
| [ChatGLM3](https://huggingface.co/THUDM) | 6B | query_key_value | chatglm3 |
| [Command-R](https://huggingface.co/CohereForAI) | 35B/104B | q_proj,v_proj | cohere |
| [DeepSeek (MoE)](https://huggingface.co/deepseek-ai) | 7B/16B/67B | q_proj,v_proj | deepseek |
| [Falcon](https://huggingface.co/tiiuae) | 7B/40B/180B | query_key_value | falcon |
| [Gemma/CodeGemma](https://huggingface.co/google) | 2B/7B | q_proj,v_proj | gemma |
| [InternLM2](https://huggingface.co/internlm) | 7B/20B | wqkv | intern2 |
| [LLaMA](https://github.com/facebookresearch/llama) | 7B/13B/33B/65B | q_proj,v_proj | - |
| [LLaMA-2](https://huggingface.co/meta-llama) | 7B/13B/70B | q_proj,v_proj | llama2 |
| [LLaMA-3](https://huggingface.co/meta-llama) | 8B/70B | q_proj,v_proj | llama3 |
| [LLaVA-1.5](https://huggingface.co/llava-hf) | 7B/13B | q_proj,v_proj | vicuna |
| [Mistral/Mixtral](https://huggingface.co/mistralai) | 7B/8x7B/8x22B | q_proj,v_proj | mistral |
| [OLMo](https://huggingface.co/allenai) | 1B/7B | q_proj,v_proj | - |
| [Phi-1.5/2](https://huggingface.co/microsoft) | 1.3B/2.7B | q_proj,v_proj | - |
| [Phi-3](https://huggingface.co/microsoft) | 3.8B | qkv_proj | phi |
| [Qwen](https://huggingface.co/Qwen) | 1.8B/7B/14B/72B | c_attn | qwen |
| [Qwen1.5 (Code/MoE)](https://huggingface.co/Qwen) | 0.5B/1.8B/4B/7B/14B/32B/72B/110B | q_proj,v_proj | qwen |
| [StarCoder2](https://huggingface.co/bigcode) | 3B/7B/15B | q_proj,v_proj | - |
| [XVERSE](https://huggingface.co/xverse) | 7B/13B/65B | q_proj,v_proj | xverse |
| [Yi](https://huggingface.co/01-ai) | 6B/9B/34B | q_proj,v_proj | yi |
| [Yuan](https://huggingface.co/IEITYuan) | 2B/51B/102B | q_proj,v_proj | yuan |
> [!NOTE]
> **Default module** is used for the `--lora_target` argument, you can use `--lora_target all` to specify all the available modules for better convergence.
>
> For the "base" models, the `--template` argument can be chosen from `default`, `alpaca`, `vicuna` etc. But make sure to use the **corresponding template** for the "instruct/chat" models.
>
> Remember to use the **SAME** template in training and inference.
Please refer to [constants.py](src/llmtuner/extras/constants.py) for a full list of models we supported.
You also can add a custom chat template to [template.py](src/llmtuner/data/template.py).
## Supported Training Approaches
| Approach | Full-tuning | Freeze-tuning | LoRA | QLoRA |
| ---------------------- | ------------------ | ------------------ | ------------------ | ------------------ |
| Pre-Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
| Supervised Fine-Tuning | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
| Reward Modeling | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
| PPO Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
| DPO Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
| ORPO Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |
## Provided Datasets
<details><summary>Pre-training datasets</summary>
- [Wiki Demo (en)](data/wiki_demo.txt)
- [RefinedWeb (en)](https://huggingface.co/datasets/tiiuae/falcon-refinedweb)
- [RedPajama V2 (en)](https://huggingface.co/datasets/togethercomputer/RedPajama-Data-V2)
- [Wikipedia (en)](https://huggingface.co/datasets/olm/olm-wikipedia-20221220)
- [Wikipedia (zh)](https://huggingface.co/datasets/pleisto/wikipedia-cn-20230720-filtered)
- [Pile (en)](https://huggingface.co/datasets/EleutherAI/pile)
- [SkyPile (zh)](https://huggingface.co/datasets/Skywork/SkyPile-150B)
- [The Stack (en)](https://huggingface.co/datasets/bigcode/the-stack)
- [StarCoder (en)](https://huggingface.co/datasets/bigcode/starcoderdata)
</details>
<details><summary>Supervised fine-tuning datasets</summary>
- [Stanford Alpaca (en)](https://github.com/tatsu-lab/stanford_alpaca)
- [Stanford Alpaca (zh)](https://github.com/ymcui/Chinese-LLaMA-Alpaca)
- [Alpaca GPT4 (en&zh)](https://github.com/Instruction-Tuning-with-GPT-4/GPT-4-LLM)
- [Self Cognition (zh)](data/self_cognition.json)
- [Open Assistant (multilingual)](https://huggingface.co/datasets/OpenAssistant/oasst1)
- [ShareGPT (zh)](https://huggingface.co/datasets/QingyiSi/Alpaca-CoT/tree/main/Chinese-instruction-collection)
- [Guanaco Dataset (multilingual)](https://huggingface.co/datasets/JosephusCheung/GuanacoDataset)
- [BELLE 2M (zh)](https://huggingface.co/datasets/BelleGroup/train_2M_CN)
- [BELLE 1M (zh)](https://huggingface.co/datasets/BelleGroup/train_1M_CN)
- [BELLE 0.5M (zh)](https://huggingface.co/datasets/BelleGroup/train_0.5M_CN)
- [BELLE Dialogue 0.4M (zh)](https://huggingface.co/datasets/BelleGroup/generated_chat_0.4M)
- [BELLE School Math 0.25M (zh)](https://huggingface.co/datasets/BelleGroup/school_math_0.25M)
- [BELLE Multiturn Chat 0.8M (zh)](https://huggingface.co/datasets/BelleGroup/multiturn_chat_0.8M)
- [UltraChat (en)](https://github.com/thunlp/UltraChat)
- [LIMA (en)](https://huggingface.co/datasets/GAIR/lima)
- [OpenPlatypus (en)](https://huggingface.co/datasets/garage-bAInd/Open-Platypus)
- [CodeAlpaca 20k (en)](https://huggingface.co/datasets/sahil2801/CodeAlpaca-20k)
- [Alpaca CoT (multilingual)](https://huggingface.co/datasets/QingyiSi/Alpaca-CoT)
- [OpenOrca (en)](https://huggingface.co/datasets/Open-Orca/OpenOrca)
- [SlimOrca (en)](https://huggingface.co/datasets/Open-Orca/SlimOrca)
- [MathInstruct (en)](https://huggingface.co/datasets/TIGER-Lab/MathInstruct)
- [Firefly 1.1M (zh)](https://huggingface.co/datasets/YeungNLP/firefly-train-1.1M)
- [Wiki QA (en)](https://huggingface.co/datasets/wiki_qa)
- [Web QA (zh)](https://huggingface.co/datasets/suolyer/webqa)
- [WebNovel (zh)](https://huggingface.co/datasets/zxbsmk/webnovel_cn)
- [Nectar (en)](https://huggingface.co/datasets/berkeley-nest/Nectar)
- [deepctrl (en&zh)](https://www.modelscope.cn/datasets/deepctrl/deepctrl-sft-data)
- [Ad Gen (zh)](https://huggingface.co/datasets/HasturOfficial/adgen)
- [ShareGPT Hyperfiltered (en)](https://huggingface.co/datasets/totally-not-an-llm/sharegpt-hyperfiltered-3k)
- [ShareGPT4 (en&zh)](https://huggingface.co/datasets/shibing624/sharegpt_gpt4)
- [UltraChat 200k (en)](https://huggingface.co/datasets/HuggingFaceH4/ultrachat_200k)
- [AgentInstruct (en)](https://huggingface.co/datasets/THUDM/AgentInstruct)
- [LMSYS Chat 1M (en)](https://huggingface.co/datasets/lmsys/lmsys-chat-1m)
- [Evol Instruct V2 (en)](https://huggingface.co/datasets/WizardLM/WizardLM_evol_instruct_V2_196k)
- [Glaive Function Calling V2 (en)](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2)
- [Cosmopedia (en)](https://huggingface.co/datasets/HuggingFaceTB/cosmopedia)
- [LLaVA mixed (en&zh)](https://huggingface.co/datasets/BUAADreamer/llava-en-zh-300k)
- [Open Assistant (de)](https://huggingface.co/datasets/mayflowergmbh/oasst_de)
- [Dolly 15k (de)](https://huggingface.co/datasets/mayflowergmbh/dolly-15k_de)
- [Alpaca GPT4 (de)](https://huggingface.co/datasets/mayflowergmbh/alpaca-gpt4_de)
- [OpenSchnabeltier (de)](https://huggingface.co/datasets/mayflowergmbh/openschnabeltier_de)
- [Evol Instruct (de)](https://huggingface.co/datasets/mayflowergmbh/evol-instruct_de)
- [Dolphin (de)](https://huggingface.co/datasets/mayflowergmbh/dolphin_de)
- [Booksum (de)](https://huggingface.co/datasets/mayflowergmbh/booksum_de)
- [Airoboros (de)](https://huggingface.co/datasets/mayflowergmbh/airoboros-3.0_de)
- [Ultrachat (de)](https://huggingface.co/datasets/mayflowergmbh/ultra-chat_de)
</details>
<details><summary>Preference datasets</summary>
- [HH-RLHF (en)](https://huggingface.co/datasets/Anthropic/hh-rlhf)
- [Open Assistant (multilingual)](https://huggingface.co/datasets/OpenAssistant/oasst1)
- [GPT-4 Generated Data (en&zh)](https://github.com/Instruction-Tuning-with-GPT-4/GPT-4-LLM)
- [Orca DPO (en)](https://huggingface.co/datasets/Intel/orca_dpo_pairs)
- [Nectar (en)](https://huggingface.co/datasets/berkeley-nest/Nectar)
- [DPO mixed (en&zh)](https://huggingface.co/datasets/hiyouga/DPO-En-Zh-20k)
- [Orca DPO (de)](https://huggingface.co/datasets/mayflowergmbh/intel_orca_dpo_pairs_de)
</details>
Some datasets require confirmation before using them, so we recommend logging in with your Hugging Face account using these commands.
```bash
pip install --upgrade huggingface_hub
huggingface-cli login
```
## Requirement
| Mandatory | Minimum | Recommend |
| ------------ | ------- | --------- |
| python | 3.8 | 3.10 |
| torch | 1.13.1 | 2.2.0 |
| transformers | 4.37.2 | 4.39.3 |
| datasets | 2.14.3 | 2.18.0 |
| accelerate | 0.27.2 | 0.28.0 |
| peft | 0.9.0 | 0.10.0 |
| trl | 0.8.1 | 0.8.1 |
| Optional | Minimum | Recommend |
| ------------ | ------- | --------- |
| CUDA | 11.6 | 12.2 |
| deepspeed | 0.10.0 | 0.14.0 |
| bitsandbytes | 0.39.0 | 0.43.0 |
| flash-attn | 2.3.0 | 2.5.6 |
### Hardware Requirement
\* *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 |
## Getting Started
### Data Preparation
Please refer to [data/README.md](data/README.md) for checking the details about the format of dataset files. You can either use datasets on HuggingFace / ModelScope hub or load the dataset in local disk.
> [!NOTE]
> Please update `data/dataset_info.json` to use your custom dataset.
### Dependence Installation
```bash
git clone https://github.com/hiyouga/LLaMA-Factory.git
conda create -n llama_factory python=3.10
conda activate llama_factory
cd LLaMA-Factory
pip install -e .[metrics]
```
Extra dependencies available: deepspeed, metrics, galore, badam, vllm, bitsandbytes, gptq, awq, aqlm, qwen, modelscope, quality
<details><summary>For Windows users</summary>
If you want to enable the quantized LoRA (QLoRA) on the Windows platform, you will be required to install a pre-built version of `bitsandbytes` library, which supports CUDA 11.1 to 12.2, please select the appropriate [release version](https://github.com/jllllll/bitsandbytes-windows-webui/releases/tag/wheels) based on your CUDA version.
```bash
pip install https://github.com/jllllll/bitsandbytes-windows-webui/releases/download/wheels/bitsandbytes-0.41.2.post2-py3-none-win_amd64.whl
```
To enable FlashAttention-2 on the Windows platform, you need to install the precompiled `flash-attn` library, which supports CUDA 12.1 to 12.2. Please download the corresponding version from [flash-attention](https://github.com/bdashore3/flash-attention/releases) based on your requirements.
</details>
### Train with LLaMA Board GUI (powered by [Gradio](https://github.com/gradio-app/gradio))
> [!IMPORTANT]
> LLaMA Board GUI only supports training on a single GPU, please use [CLI](#command-line-interface) for distributed training.
#### Use local environment
```bash
export CUDA_VISIBLE_DEVICES=0 # `set CUDA_VISIBLE_DEVICES=0` for Windows
export GRADIO_SERVER_PORT=7860 # `set GRADIO_SERVER_PORT=7860` for Windows
python src/train_web.py # or python -m llmtuner.webui.interface
```
<details><summary>For Alibaba Cloud users</summary>
If you encountered display problems in LLaMA Board on Alibaba Cloud, try using the following command to set environment variables before starting LLaMA Board:
```bash
export GRADIO_ROOT_PATH=/${JUPYTER_NAME}/proxy/7860/
```
</details>
#### Use Docker
```bash
docker build -f ./Dockerfile -t llama-factory:latest .
docker run --gpus=all \
-v ./hf_cache:/root/.cache/huggingface/ \
-v ./data:/app/data \
-v ./output:/app/output \
-e CUDA_VISIBLE_DEVICES=0 \
-p 7860:7860 \
--shm-size 16G \
--name llama_factory \
-d llama-factory:latest
```
#### Use Docker Compose
```bash
docker compose -f ./docker-compose.yml up -d
```
<details><summary>Details about volume</summary>
- hf_cache: Utilize Hugging Face cache on the host machine. Reassignable if a cache already exists in a different directory.
- data: Place datasets on this dir of the host machine so that they can be selected on LLaMA Board GUI.
- output: Set export dir to this location so that the merged result can be accessed directly on the host machine.
</details>
### Train with Command Line Interface
See [examples/README.md](examples/README.md) for usage.
Use `python src/train_bash.py -h` to display arguments description.
### Deploy with OpenAI-style API and vLLM
```bash
CUDA_VISIBLE_DEVICES=0,1 API_PORT=8000 python src/api_demo.py \
--model_name_or_path meta-llama/Meta-Llama-3-8B-Instruct \
--template llama3 \
--infer_backend vllm \
--vllm_enforce_eager
```
### Download from ModelScope Hub
If you have trouble with downloading models and datasets from Hugging Face, you can use ModelScope.
```bash
export USE_MODELSCOPE_HUB=1 # `set USE_MODELSCOPE_HUB=1` for Windows
```
Train the model by specifying a model ID of the ModelScope Hub as the `--model_name_or_path`. You can find a full list of model IDs at [ModelScope Hub](https://modelscope.cn/models), e.g., `LLM-Research/Meta-Llama-3-8B-Instruct`.
## Projects using LLaMA Factory
If you have a project that should be incorporated, please contact via email or create a pull request.
<details><summary>Click to show</summary>
1. Wang et al. ESRL: Efficient Sampling-based Reinforcement Learning for Sequence Generation. 2023. [[arxiv]](https://arxiv.org/abs/2308.02223)
1. Yu et al. Open, Closed, or Small Language Models for Text Classification? 2023. [[arxiv]](https://arxiv.org/abs/2308.10092)
1. Wang et al. UbiPhysio: Support Daily Functioning, Fitness, and Rehabilitation with Action Understanding and Feedback in Natural Language. 2023. [[arxiv]](https://arxiv.org/abs/2308.10526)
1. Luceri et al. Leveraging Large Language Models to Detect Influence Campaigns in Social Media. 2023. [[arxiv]](https://arxiv.org/abs/2311.07816)
1. Zhang et al. Alleviating Hallucinations of Large Language Models through Induced Hallucinations. 2023. [[arxiv]](https://arxiv.org/abs/2312.15710)
1. Wang et al. Know Your Needs Better: Towards Structured Understanding of Marketer Demands with Analogical Reasoning Augmented LLMs. 2024. [[arxiv]](https://arxiv.org/abs/2401.04319)
1. Wang et al. CANDLE: Iterative Conceptualization and Instantiation Distillation from Large Language Models for Commonsense Reasoning. 2024. [[arxiv]](https://arxiv.org/abs/2401.07286)
1. Choi et al. FACT-GPT: Fact-Checking Augmentation via Claim Matching with LLMs. 2024. [[arxiv]](https://arxiv.org/abs/2402.05904)
1. Zhang et al. AutoMathText: Autonomous Data Selection with Language Models for Mathematical Texts. 2024. [[arxiv]](https://arxiv.org/abs/2402.07625)
1. Lyu et al. KnowTuning: Knowledge-aware Fine-tuning for Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.11176)
1. Yang et al. LaCo: Large Language Model Pruning via Layer Collaps. 2024. [[arxiv]](https://arxiv.org/abs/2402.11187)
1. Bhardwaj et al. Language Models are Homer Simpson! Safety Re-Alignment of Fine-tuned Language Models through Task Arithmetic. 2024. [[arxiv]](https://arxiv.org/abs/2402.11746)
1. Yang et al. Enhancing Empathetic Response Generation by Augmenting LLMs with Small-scale Empathetic Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.11801)
1. Yi et al. Generation Meets Verification: Accelerating Large Language Model Inference with Smart Parallel Auto-Correct Decoding. 2024. [[arxiv]](https://arxiv.org/abs/2402.11809)
1. Cao et al. Head-wise Shareable Attention for Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.11819)
1. Zhang et al. Enhancing Multilingual Capabilities of Large Language Models through Self-Distillation from Resource-Rich Languages. 2024. [[arxiv]](https://arxiv.org/abs/2402.12204)
1. Kim et al. Efficient and Effective Vocabulary Expansion Towards Multilingual Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.14714)
1. Yu et al. KIEval: A Knowledge-grounded Interactive Evaluation Framework for Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.15043)
1. Huang et al. Key-Point-Driven Data Synthesis with its Enhancement on Mathematical Reasoning. 2024. [[arxiv]](https://arxiv.org/abs/2403.02333)
1. Duan et al. Negating Negatives: Alignment without Human Positive Samples via Distributional Dispreference Optimization. 2024. [[arxiv]](https://arxiv.org/abs/2403.03419)
1. Xie and Schwertfeger. Empowering Robotics with Large Language Models: osmAG Map Comprehension with LLMs. 2024. [[arxiv]](https://arxiv.org/abs/2403.08228)
1. Zhang et al. EDT: Improving Large Language Models' Generation by Entropy-based Dynamic Temperature Sampling. 2024. [[arxiv]](https://arxiv.org/abs/2403.14541)
1. Weller et al. FollowIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions. 2024. [[arxiv]](https://arxiv.org/abs/2403.15246)
1. Hongbin Na. CBT-LLM: A Chinese Large Language Model for Cognitive Behavioral Therapy-based Mental Health Question Answering. 2024. [[arxiv]](https://arxiv.org/abs/2403.16008)
1. Zan et al. CodeS: Natural Language to Code Repository via Multi-Layer Sketch. 2024. [[arxiv]](https://arxiv.org/abs/2403.16443)
1. Liu et al. Extensive Self-Contrast Enables Feedback-Free Language Model Alignment. 2024. [[arxiv]](https://arxiv.org/abs/2404.00604)
1. Luo et al. BAdam: A Memory Efficient Full Parameter Training Method for Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2404.02827)
1. Du et al. Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model. 2024. [[arxiv]](https://arxiv.org/abs/2404.04167)
1. Liu et al. Dynamic Generation of Personalities with Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2404.07084)
1. **[StarWhisper](https://github.com/Yu-Yang-Li/StarWhisper)**: A large language model for Astronomy, based on ChatGLM2-6B and Qwen-14B.
1. **[DISC-LawLLM](https://github.com/FudanDISC/DISC-LawLLM)**: A large language model specialized in Chinese legal domain, based on Baichuan-13B, is capable of retrieving and reasoning on legal knowledge.
1. **[Sunsimiao](https://github.com/thomas-yanxin/Sunsimiao)**: A large language model specialized in Chinese medical domain, based on Baichuan-7B and ChatGLM-6B.
1. **[CareGPT](https://github.com/WangRongsheng/CareGPT)**: A series of large language models for Chinese medical domain, based on LLaMA2-7B and Baichuan-13B.
1. **[MachineMindset](https://github.com/PKU-YuanGroup/Machine-Mindset/)**: A series of MBTI Personality large language models, capable of giving any LLM 16 different personality types based on different datasets and training methods.
</details>
## License
This repository is licensed under the [Apache-2.0 License](LICENSE).
Please follow the model licenses to use the corresponding model weights: [Baichuan2](https://huggingface.co/baichuan-inc/Baichuan2-7B-Base/blob/main/Community%20License%20for%20Baichuan%202%20Model.pdf) / [BLOOM](https://huggingface.co/spaces/bigscience/license) / [ChatGLM3](https://github.com/THUDM/ChatGLM3/blob/main/MODEL_LICENSE) / [Command-R](https://cohere.com/c4ai-cc-by-nc-license) / [DeepSeek](https://github.com/deepseek-ai/DeepSeek-LLM/blob/main/LICENSE-MODEL) / [Falcon](https://huggingface.co/tiiuae/falcon-180B/blob/main/LICENSE.txt) / [Gemma](https://ai.google.dev/gemma/terms) / [InternLM2](https://github.com/InternLM/InternLM#license) / [LLaMA](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md) / [LLaMA-2/LLaVA-1.5](https://ai.meta.com/llama/license/) / [LLaMA-3](https://llama.meta.com/llama3/license/) / [Mistral](LICENSE) / [OLMo](LICENSE) / [Phi-1.5/2](https://huggingface.co/microsoft/phi-1_5/resolve/main/Research%20License.docx) / [Phi-3](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/blob/main/LICENSE) / [Qwen](https://github.com/QwenLM/Qwen/blob/main/Tongyi%20Qianwen%20LICENSE%20AGREEMENT) / [StarCoder2](https://huggingface.co/spaces/bigcode/bigcode-model-license-agreement) / [XVERSE](https://github.com/xverse-ai/XVERSE-13B/blob/main/MODEL_LICENSE.pdf) / [Yi](https://huggingface.co/01-ai/Yi-6B/blob/main/LICENSE) / [Yuan](https://github.com/IEIT-Yuan/Yuan-2.0/blob/main/LICENSE-Yuan)
## Citation
If this work is helpful, please kindly cite as:
```bibtex
@article{zheng2024llamafactory,
title={LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models},
author={Yaowei Zheng and Richong Zhang and Junhao Zhang and Yanhan Ye and Zheyan Luo and Yongqiang Ma},
journal={arXiv preprint arXiv:2403.13372},
year={2024},
url={http://arxiv.org/abs/2403.13372}
}
```
## Acknowledgement
This repo benefits from [PEFT](https://github.com/huggingface/peft), [TRL](https://github.com/huggingface/trl), [QLoRA](https://github.com/artidoro/qlora) and [FastChat](https://github.com/lm-sys/FastChat). Thanks for their wonderful works.
## Star History
Raw data
{
"_id": null,
"home_page": "https://github.com/hiyouga/LLaMA-Factory",
"name": "llmtuner",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8.0",
"maintainer_email": null,
"keywords": "LLaMA, BLOOM, Falcon, LLM, ChatGPT, transformer, pytorch, deep learning",
"author": "hiyouga",
"author_email": "hiyouga@buaa.edu.cn",
"download_url": "https://files.pythonhosted.org/packages/e0/09/140b915097765b07e8e3e01ca9c433722892be4d0437e91803f1b2de848c/llmtuner-0.7.0.tar.gz",
"platform": null,
"description": "\n\n[](https://github.com/hiyouga/LLaMA-Factory/stargazers)\n[](LICENSE)\n[](https://github.com/hiyouga/LLaMA-Factory/commits/main)\n[](https://pypi.org/project/llmtuner/)\n[](https://pypi.org/project/llmtuner/)\n[](#projects-using-llama-factory)\n[](https://github.com/hiyouga/LLaMA-Factory/pulls)\n[](https://discord.gg/rKfvV9r9FK)\n[](https://twitter.com/llamafactory_ai)\n[](https://huggingface.co/spaces/hiyouga/LLaMA-Board)\n[](https://modelscope.cn/studios/hiyouga/LLaMA-Board)\n[](https://colab.research.google.com/drive/1eRTPn37ltBbYsISy9Aw2NuI2Aq5CQrD9?usp=sharing)\n\n\ud83d\udc4b Join our [WeChat](assets/wechat.jpg).\n\n\\[ English | [\u4e2d\u6587](README_zh.md) \\]\n\n**Fine-tuning a large language model can be easy as...**\n\nhttps://github.com/hiyouga/LLaMA-Factory/assets/16256802/9840a653-7e9c-41c8-ae89-7ace5698baf6\n\nChoose your path:\n\n- **Colab**: https://colab.research.google.com/drive/1eRTPn37ltBbYsISy9Aw2NuI2Aq5CQrD9?usp=sharing\n- **Local machine**: Please refer to [usage](#getting-started)\n\n## Table of Contents\n\n- [Features](#features)\n- [Benchmark](#benchmark)\n- [Changelog](#changelog)\n- [Supported Models](#supported-models)\n- [Supported Training Approaches](#supported-training-approaches)\n- [Provided Datasets](#provided-datasets)\n- [Requirement](#requirement)\n- [Getting Started](#getting-started)\n- [Projects using LLaMA Factory](#projects-using-llama-factory)\n- [License](#license)\n- [Citation](#citation)\n- [Acknowledgement](#acknowledgement)\n\n## Features\n\n- **Various models**: LLaMA, LLaVA, Mistral, Mixtral-MoE, Qwen, Yi, Gemma, Baichuan, ChatGLM, Phi, etc.\n- **Integrated methods**: (Continuous) pre-training, (multimodal) supervised fine-tuning, reward modeling, PPO, DPO and ORPO.\n- **Scalable resources**: 32-bit full-tuning, 16-bit freeze-tuning, 16-bit LoRA and 2/4/8-bit QLoRA via AQLM/AWQ/GPTQ/LLM.int8.\n- **Advanced algorithms**: GaLore, BAdam, DoRA, LongLoRA, LLaMA Pro, Mixture-of-Depths, LoRA+, LoftQ and Agent tuning.\n- **Practical tricks**: FlashAttention-2, Unsloth, RoPE scaling, NEFTune and rsLoRA.\n- **Experiment monitors**: LlamaBoard, TensorBoard, Wandb, MLflow, etc.\n- **Faster inference**: OpenAI-style API, Gradio UI and CLI with vLLM worker.\n\n## Benchmark\n\nCompared to ChatGLM's [P-Tuning](https://github.com/THUDM/ChatGLM2-6B/tree/main/ptuning), LLaMA Factory's LoRA tuning offers up to **3.7 times faster** training speed with a better Rouge score on the advertising text generation task. By leveraging 4-bit quantization technique, LLaMA Factory's QLoRA further improves the efficiency regarding the GPU memory.\n\n\n\n<details><summary>Definitions</summary>\n\n- **Training Speed**: the number of training samples processed per second during the training. (bs=4, cutoff_len=1024)\n- **Rouge Score**: Rouge-2 score on the development set of the [advertising text generation](https://aclanthology.org/D19-1321.pdf) task. (bs=4, cutoff_len=1024)\n- **GPU Memory**: Peak GPU memory usage in 4-bit quantized training. (bs=1, cutoff_len=1024)\n- We adopt `pre_seq_len=128` for ChatGLM's P-Tuning and `lora_rank=32` for LLaMA Factory's LoRA tuning.\n\n</details>\n\n## Changelog\n\n[24/04/26] We supported fine-tuning the **LLaVA-1.5** multimodal LLMs. See `examples/lora_single_gpu/sft_mllm.sh` for usage.\n\n[24/04/22] We provided a **[Colab notebook](https://colab.research.google.com/drive/1eRTPn37ltBbYsISy9Aw2NuI2Aq5CQrD9?usp=sharing)** for fine-tuning the Llama-3 model on a free T4 GPU. Two Llama-3-derived models fine-tuned using LLaMA Factory are available at Hugging Face, check [Llama3-8B-Chinese-Chat](https://huggingface.co/shenzhi-wang/Llama3-8B-Chinese-Chat) and [Llama3-Chinese](https://huggingface.co/zhichen/Llama3-Chinese) for details.\n\n[24/04/21] We supported **[Mixture-of-Depths](https://arxiv.org/abs/2404.02258)** according to [AstraMindAI's implementation](https://github.com/astramind-ai/Mixture-of-depths). See `examples/extras/mod` for usage.\n\n[24/04/16] We supported **[BAdam](https://arxiv.org/abs/2404.02827)**. See `examples/extras/badam` for usage.\n\n[24/04/16] We supported **[unsloth](https://github.com/unslothai/unsloth)**'s long-sequence training (Llama-2-7B-56k within 24GB). It achieves **117%** speed and **50%** memory compared with FlashAttention-2, more benchmarks can be found in [this page](https://github.com/hiyouga/LLaMA-Factory/wiki/Performance-comparison).\n\n<details><summary>Full Changelog</summary>\n\n[24/03/31] We supported **[ORPO](https://arxiv.org/abs/2403.07691)**. See `examples/lora_single_gpu` for usage.\n\n[24/03/21] Our paper \"[LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://arxiv.org/abs/2403.13372)\" is available at arXiv!\n\n[24/03/20] We supported **FSDP+QLoRA** that fine-tunes a 70B model on 2x24GB GPUs. See `examples/extras/fsdp_qlora` for usage.\n\n[24/03/13] We supported **[LoRA+](https://arxiv.org/abs/2402.12354)**. See `examples/extras/loraplus` for usage.\n\n[24/03/07] We supported gradient low-rank projection (**[GaLore](https://arxiv.org/abs/2403.03507)**) algorithm. See `examples/extras/galore` for usage.\n\n[24/03/07] We integrated **[vLLM](https://github.com/vllm-project/vllm)** for faster and concurrent inference. Try `--infer_backend vllm` to enjoy **270%** inference speed. (LoRA is not yet supported, merge it first.)\n\n[24/02/28] We supported weight-decomposed LoRA (**[DoRA](https://arxiv.org/abs/2402.09353)**). Try `--use_dora` to activate DoRA training.\n\n[24/02/15] We supported **block expansion** proposed by [LLaMA Pro](https://github.com/TencentARC/LLaMA-Pro). See `examples/extras/llama_pro` for usage.\n\n[24/02/05] Qwen1.5 (Qwen2 beta version) series models are supported in LLaMA-Factory. Check this [blog post](https://qwenlm.github.io/blog/qwen1.5/) for details.\n\n[24/01/18] We supported **agent tuning** for most models, equipping model with tool using abilities by fine-tuning with `--dataset glaive_toolcall`.\n\n[23/12/23] We supported **[unsloth](https://github.com/unslothai/unsloth)**'s implementation to boost LoRA tuning for the LLaMA, Mistral and Yi models. Try `--use_unsloth` argument to activate unsloth patch. It achieves **170%** speed in our benchmark, check [this page](https://github.com/hiyouga/LLaMA-Factory/wiki/Performance-comparison) for details.\n\n[23/12/12] We supported fine-tuning the latest MoE model **[Mixtral 8x7B](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1)** in our framework. See hardware requirement [here](#hardware-requirement).\n\n[23/12/01] We supported downloading pre-trained models and datasets from the **[ModelScope Hub](https://modelscope.cn/models)** for Chinese mainland users. See [this tutorial](#use-modelscope-hub-optional) for usage.\n\n[23/10/21] We supported **[NEFTune](https://arxiv.org/abs/2310.05914)** trick for fine-tuning. Try `--neftune_noise_alpha` argument to activate NEFTune, e.g., `--neftune_noise_alpha 5`.\n\n[23/09/27] We supported **$S^2$-Attn** proposed by [LongLoRA](https://github.com/dvlab-research/LongLoRA) for the LLaMA models. Try `--shift_attn` argument to enable shift short attention.\n\n[23/09/23] We integrated MMLU, C-Eval and CMMLU benchmarks in this repo. See [this example](#evaluation) to evaluate your models.\n\n[23/09/10] We supported **[FlashAttention-2](https://github.com/Dao-AILab/flash-attention)**. Try `--flash_attn fa2` argument to enable FlashAttention-2 if you are using RTX4090, A100 or H100 GPUs.\n\n[23/08/12] We supported **RoPE scaling** to extend the context length of the LLaMA models. Try `--rope_scaling linear` argument in training and `--rope_scaling dynamic` argument at inference to extrapolate the position embeddings.\n\n[23/08/11] We supported **[DPO training](https://arxiv.org/abs/2305.18290)** for instruction-tuned models. See [this example](#dpo-training) to train your models.\n\n[23/07/31] We supported **dataset streaming**. Try `--streaming` and `--max_steps 10000` arguments to load your dataset in streaming mode.\n\n[23/07/29] We released two instruction-tuned 13B models at Hugging Face. See these Hugging Face Repos ([LLaMA-2](https://huggingface.co/hiyouga/Llama-2-Chinese-13b-chat) / [Baichuan](https://huggingface.co/hiyouga/Baichuan-13B-sft)) for details.\n\n[23/07/18] We developed an **all-in-one Web UI** for training, evaluation and inference. Try `train_web.py` to fine-tune models in your Web browser. Thank [@KanadeSiina](https://github.com/KanadeSiina) and [@codemayq](https://github.com/codemayq) for their efforts in the development.\n\n[23/07/09] We released **[FastEdit](https://github.com/hiyouga/FastEdit)** \u26a1\ud83e\ude79, an easy-to-use package for editing the factual knowledge of large language models efficiently. Please follow [FastEdit](https://github.com/hiyouga/FastEdit) if you are interested.\n\n[23/06/29] We provided a **reproducible example** of training a chat model using instruction-following datasets, see [Baichuan-7B-sft](https://huggingface.co/hiyouga/Baichuan-7B-sft) for details.\n\n[23/06/22] We aligned the [demo API](src/api_demo.py) with the [OpenAI's](https://platform.openai.com/docs/api-reference/chat) format where you can insert the fine-tuned model in **arbitrary ChatGPT-based applications**.\n\n[23/06/03] We supported quantized training and inference (aka **[QLoRA](https://github.com/artidoro/qlora)**). Try `--quantization_bit 4/8` argument to work with quantized models.\n\n</details>\n\n## Supported Models\n\n| Model | Model size | Default module | Template |\n| -------------------------------------------------------- | -------------------------------- | ----------------- | --------- |\n| [Baichuan2](https://huggingface.co/baichuan-inc) | 7B/13B | W_pack | baichuan2 |\n| [BLOOM](https://huggingface.co/bigscience) | 560M/1.1B/1.7B/3B/7.1B/176B | query_key_value | - |\n| [BLOOMZ](https://huggingface.co/bigscience) | 560M/1.1B/1.7B/3B/7.1B/176B | query_key_value | - |\n| [ChatGLM3](https://huggingface.co/THUDM) | 6B | query_key_value | chatglm3 |\n| [Command-R](https://huggingface.co/CohereForAI) | 35B/104B | q_proj,v_proj | cohere |\n| [DeepSeek (MoE)](https://huggingface.co/deepseek-ai) | 7B/16B/67B | q_proj,v_proj | deepseek |\n| [Falcon](https://huggingface.co/tiiuae) | 7B/40B/180B | query_key_value | falcon |\n| [Gemma/CodeGemma](https://huggingface.co/google) | 2B/7B | q_proj,v_proj | gemma |\n| [InternLM2](https://huggingface.co/internlm) | 7B/20B | wqkv | intern2 |\n| [LLaMA](https://github.com/facebookresearch/llama) | 7B/13B/33B/65B | q_proj,v_proj | - |\n| [LLaMA-2](https://huggingface.co/meta-llama) | 7B/13B/70B | q_proj,v_proj | llama2 |\n| [LLaMA-3](https://huggingface.co/meta-llama) | 8B/70B | q_proj,v_proj | llama3 |\n| [LLaVA-1.5](https://huggingface.co/llava-hf) | 7B/13B | q_proj,v_proj | vicuna |\n| [Mistral/Mixtral](https://huggingface.co/mistralai) | 7B/8x7B/8x22B | q_proj,v_proj | mistral |\n| [OLMo](https://huggingface.co/allenai) | 1B/7B | q_proj,v_proj | - |\n| [Phi-1.5/2](https://huggingface.co/microsoft) | 1.3B/2.7B | q_proj,v_proj | - |\n| [Phi-3](https://huggingface.co/microsoft) | 3.8B | qkv_proj | phi |\n| [Qwen](https://huggingface.co/Qwen) | 1.8B/7B/14B/72B | c_attn | qwen |\n| [Qwen1.5 (Code/MoE)](https://huggingface.co/Qwen) | 0.5B/1.8B/4B/7B/14B/32B/72B/110B | q_proj,v_proj | qwen |\n| [StarCoder2](https://huggingface.co/bigcode) | 3B/7B/15B | q_proj,v_proj | - |\n| [XVERSE](https://huggingface.co/xverse) | 7B/13B/65B | q_proj,v_proj | xverse |\n| [Yi](https://huggingface.co/01-ai) | 6B/9B/34B | q_proj,v_proj | yi |\n| [Yuan](https://huggingface.co/IEITYuan) | 2B/51B/102B | q_proj,v_proj | yuan |\n\n> [!NOTE]\n> **Default module** is used for the `--lora_target` argument, you can use `--lora_target all` to specify all the available modules for better convergence.\n>\n> For the \"base\" models, the `--template` argument can be chosen from `default`, `alpaca`, `vicuna` etc. But make sure to use the **corresponding template** for the \"instruct/chat\" models.\n>\n> Remember to use the **SAME** template in training and inference.\n\nPlease refer to [constants.py](src/llmtuner/extras/constants.py) for a full list of models we supported.\n\nYou also can add a custom chat template to [template.py](src/llmtuner/data/template.py).\n\n## Supported Training Approaches\n\n| Approach | Full-tuning | Freeze-tuning | LoRA | QLoRA |\n| ---------------------- | ------------------ | ------------------ | ------------------ | ------------------ |\n| Pre-Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |\n| Supervised Fine-Tuning | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |\n| Reward Modeling | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |\n| PPO Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |\n| DPO Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |\n| ORPO Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: |\n\n## Provided Datasets\n\n<details><summary>Pre-training datasets</summary>\n\n- [Wiki Demo (en)](data/wiki_demo.txt)\n- [RefinedWeb (en)](https://huggingface.co/datasets/tiiuae/falcon-refinedweb)\n- [RedPajama V2 (en)](https://huggingface.co/datasets/togethercomputer/RedPajama-Data-V2)\n- [Wikipedia (en)](https://huggingface.co/datasets/olm/olm-wikipedia-20221220)\n- [Wikipedia (zh)](https://huggingface.co/datasets/pleisto/wikipedia-cn-20230720-filtered)\n- [Pile (en)](https://huggingface.co/datasets/EleutherAI/pile)\n- [SkyPile (zh)](https://huggingface.co/datasets/Skywork/SkyPile-150B)\n- [The Stack (en)](https://huggingface.co/datasets/bigcode/the-stack)\n- [StarCoder (en)](https://huggingface.co/datasets/bigcode/starcoderdata)\n\n</details>\n\n<details><summary>Supervised fine-tuning datasets</summary>\n\n- [Stanford Alpaca (en)](https://github.com/tatsu-lab/stanford_alpaca)\n- [Stanford Alpaca (zh)](https://github.com/ymcui/Chinese-LLaMA-Alpaca)\n- [Alpaca GPT4 (en&zh)](https://github.com/Instruction-Tuning-with-GPT-4/GPT-4-LLM)\n- [Self Cognition (zh)](data/self_cognition.json)\n- [Open Assistant (multilingual)](https://huggingface.co/datasets/OpenAssistant/oasst1)\n- [ShareGPT (zh)](https://huggingface.co/datasets/QingyiSi/Alpaca-CoT/tree/main/Chinese-instruction-collection)\n- [Guanaco Dataset (multilingual)](https://huggingface.co/datasets/JosephusCheung/GuanacoDataset)\n- [BELLE 2M (zh)](https://huggingface.co/datasets/BelleGroup/train_2M_CN)\n- [BELLE 1M (zh)](https://huggingface.co/datasets/BelleGroup/train_1M_CN)\n- [BELLE 0.5M (zh)](https://huggingface.co/datasets/BelleGroup/train_0.5M_CN)\n- [BELLE Dialogue 0.4M (zh)](https://huggingface.co/datasets/BelleGroup/generated_chat_0.4M)\n- [BELLE School Math 0.25M (zh)](https://huggingface.co/datasets/BelleGroup/school_math_0.25M)\n- [BELLE Multiturn Chat 0.8M (zh)](https://huggingface.co/datasets/BelleGroup/multiturn_chat_0.8M)\n- [UltraChat (en)](https://github.com/thunlp/UltraChat)\n- [LIMA (en)](https://huggingface.co/datasets/GAIR/lima)\n- [OpenPlatypus (en)](https://huggingface.co/datasets/garage-bAInd/Open-Platypus)\n- [CodeAlpaca 20k (en)](https://huggingface.co/datasets/sahil2801/CodeAlpaca-20k)\n- [Alpaca CoT (multilingual)](https://huggingface.co/datasets/QingyiSi/Alpaca-CoT)\n- [OpenOrca (en)](https://huggingface.co/datasets/Open-Orca/OpenOrca)\n- [SlimOrca (en)](https://huggingface.co/datasets/Open-Orca/SlimOrca)\n- [MathInstruct (en)](https://huggingface.co/datasets/TIGER-Lab/MathInstruct)\n- [Firefly 1.1M (zh)](https://huggingface.co/datasets/YeungNLP/firefly-train-1.1M)\n- [Wiki QA (en)](https://huggingface.co/datasets/wiki_qa)\n- [Web QA (zh)](https://huggingface.co/datasets/suolyer/webqa)\n- [WebNovel (zh)](https://huggingface.co/datasets/zxbsmk/webnovel_cn)\n- [Nectar (en)](https://huggingface.co/datasets/berkeley-nest/Nectar)\n- [deepctrl (en&zh)](https://www.modelscope.cn/datasets/deepctrl/deepctrl-sft-data)\n- [Ad Gen (zh)](https://huggingface.co/datasets/HasturOfficial/adgen)\n- [ShareGPT Hyperfiltered (en)](https://huggingface.co/datasets/totally-not-an-llm/sharegpt-hyperfiltered-3k)\n- [ShareGPT4 (en&zh)](https://huggingface.co/datasets/shibing624/sharegpt_gpt4)\n- [UltraChat 200k (en)](https://huggingface.co/datasets/HuggingFaceH4/ultrachat_200k)\n- [AgentInstruct (en)](https://huggingface.co/datasets/THUDM/AgentInstruct)\n- [LMSYS Chat 1M (en)](https://huggingface.co/datasets/lmsys/lmsys-chat-1m)\n- [Evol Instruct V2 (en)](https://huggingface.co/datasets/WizardLM/WizardLM_evol_instruct_V2_196k)\n- [Glaive Function Calling V2 (en)](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2)\n- [Cosmopedia (en)](https://huggingface.co/datasets/HuggingFaceTB/cosmopedia)\n- [LLaVA mixed (en&zh)](https://huggingface.co/datasets/BUAADreamer/llava-en-zh-300k)\n- [Open Assistant (de)](https://huggingface.co/datasets/mayflowergmbh/oasst_de)\n- [Dolly 15k (de)](https://huggingface.co/datasets/mayflowergmbh/dolly-15k_de)\n- [Alpaca GPT4 (de)](https://huggingface.co/datasets/mayflowergmbh/alpaca-gpt4_de)\n- [OpenSchnabeltier (de)](https://huggingface.co/datasets/mayflowergmbh/openschnabeltier_de)\n- [Evol Instruct (de)](https://huggingface.co/datasets/mayflowergmbh/evol-instruct_de)\n- [Dolphin (de)](https://huggingface.co/datasets/mayflowergmbh/dolphin_de)\n- [Booksum (de)](https://huggingface.co/datasets/mayflowergmbh/booksum_de)\n- [Airoboros (de)](https://huggingface.co/datasets/mayflowergmbh/airoboros-3.0_de)\n- [Ultrachat (de)](https://huggingface.co/datasets/mayflowergmbh/ultra-chat_de)\n\n</details>\n\n<details><summary>Preference datasets</summary>\n\n- [HH-RLHF (en)](https://huggingface.co/datasets/Anthropic/hh-rlhf)\n- [Open Assistant (multilingual)](https://huggingface.co/datasets/OpenAssistant/oasst1)\n- [GPT-4 Generated Data (en&zh)](https://github.com/Instruction-Tuning-with-GPT-4/GPT-4-LLM)\n- [Orca DPO (en)](https://huggingface.co/datasets/Intel/orca_dpo_pairs)\n- [Nectar (en)](https://huggingface.co/datasets/berkeley-nest/Nectar)\n- [DPO mixed (en&zh)](https://huggingface.co/datasets/hiyouga/DPO-En-Zh-20k)\n- [Orca DPO (de)](https://huggingface.co/datasets/mayflowergmbh/intel_orca_dpo_pairs_de)\n\n</details>\n\nSome datasets require confirmation before using them, so we recommend logging in with your Hugging Face account using these commands.\n\n```bash\npip install --upgrade huggingface_hub\nhuggingface-cli login\n```\n\n## Requirement\n\n| Mandatory | Minimum | Recommend |\n| ------------ | ------- | --------- |\n| python | 3.8 | 3.10 |\n| torch | 1.13.1 | 2.2.0 |\n| transformers | 4.37.2 | 4.39.3 |\n| datasets | 2.14.3 | 2.18.0 |\n| accelerate | 0.27.2 | 0.28.0 |\n| peft | 0.9.0 | 0.10.0 |\n| trl | 0.8.1 | 0.8.1 |\n\n| Optional | Minimum | Recommend |\n| ------------ | ------- | --------- |\n| CUDA | 11.6 | 12.2 |\n| deepspeed | 0.10.0 | 0.14.0 |\n| bitsandbytes | 0.39.0 | 0.43.0 |\n| flash-attn | 2.3.0 | 2.5.6 |\n\n### Hardware Requirement\n\n\\* *estimated*\n\n| Method | Bits | 7B | 13B | 30B | 70B | 110B | 8x7B | 8x22B |\n| ----------------- | ---- | ----- | ----- | ----- | ------ | ------ | ----- | ------ |\n| Full | AMP | 120GB | 240GB | 600GB | 1200GB | 2000GB | 900GB | 2400GB |\n| Full | 16 | 60GB | 120GB | 300GB | 600GB | 900GB | 400GB | 1200GB |\n| Freeze | 16 | 20GB | 40GB | 80GB | 200GB | 360GB | 160GB | 400GB |\n| LoRA/GaLore/BAdam | 16 | 16GB | 32GB | 64GB | 160GB | 240GB | 120GB | 320GB |\n| QLoRA | 8 | 10GB | 20GB | 40GB | 80GB | 140GB | 60GB | 160GB |\n| QLoRA | 4 | 6GB | 12GB | 24GB | 48GB | 72GB | 30GB | 96GB |\n| QLoRA | 2 | 4GB | 8GB | 16GB | 24GB | 48GB | 18GB | 48GB |\n\n## Getting Started\n\n### Data Preparation\n\nPlease refer to [data/README.md](data/README.md) for checking the details about the format of dataset files. You can either use datasets on HuggingFace / ModelScope hub or load the dataset in local disk.\n\n> [!NOTE]\n> Please update `data/dataset_info.json` to use your custom dataset.\n\n### Dependence Installation\n\n```bash\ngit clone https://github.com/hiyouga/LLaMA-Factory.git\nconda create -n llama_factory python=3.10\nconda activate llama_factory\ncd LLaMA-Factory\npip install -e .[metrics]\n```\n\nExtra dependencies available: deepspeed, metrics, galore, badam, vllm, bitsandbytes, gptq, awq, aqlm, qwen, modelscope, quality\n\n<details><summary>For Windows users</summary>\n\nIf you want to enable the quantized LoRA (QLoRA) on the Windows platform, you will be required to install a pre-built version of `bitsandbytes` library, which supports CUDA 11.1 to 12.2, please select the appropriate [release version](https://github.com/jllllll/bitsandbytes-windows-webui/releases/tag/wheels) based on your CUDA version.\n\n```bash\npip install https://github.com/jllllll/bitsandbytes-windows-webui/releases/download/wheels/bitsandbytes-0.41.2.post2-py3-none-win_amd64.whl\n```\n\nTo enable FlashAttention-2 on the Windows platform, you need to install the precompiled `flash-attn` library, which supports CUDA 12.1 to 12.2. Please download the corresponding version from [flash-attention](https://github.com/bdashore3/flash-attention/releases) based on your requirements.\n\n</details>\n\n### Train with LLaMA Board GUI (powered by [Gradio](https://github.com/gradio-app/gradio))\n\n> [!IMPORTANT]\n> LLaMA Board GUI only supports training on a single GPU, please use [CLI](#command-line-interface) for distributed training.\n\n#### Use local environment\n\n```bash\nexport CUDA_VISIBLE_DEVICES=0 # `set CUDA_VISIBLE_DEVICES=0` for Windows\nexport GRADIO_SERVER_PORT=7860 # `set GRADIO_SERVER_PORT=7860` for Windows\npython src/train_web.py # or python -m llmtuner.webui.interface\n```\n\n<details><summary>For Alibaba Cloud users</summary>\n\nIf you encountered display problems in LLaMA Board on Alibaba Cloud, try using the following command to set environment variables before starting LLaMA Board:\n\n```bash\nexport GRADIO_ROOT_PATH=/${JUPYTER_NAME}/proxy/7860/\n```\n\n</details>\n\n#### Use Docker\n\n```bash\ndocker build -f ./Dockerfile -t llama-factory:latest .\ndocker run --gpus=all \\\n -v ./hf_cache:/root/.cache/huggingface/ \\\n -v ./data:/app/data \\\n -v ./output:/app/output \\\n -e CUDA_VISIBLE_DEVICES=0 \\\n -p 7860:7860 \\\n --shm-size 16G \\\n --name llama_factory \\\n -d llama-factory:latest\n```\n\n#### Use Docker Compose\n\n```bash\ndocker compose -f ./docker-compose.yml up -d\n```\n\n<details><summary>Details about volume</summary>\n\n- hf_cache: Utilize Hugging Face cache on the host machine. Reassignable if a cache already exists in a different directory.\n- data: Place datasets on this dir of the host machine so that they can be selected on LLaMA Board GUI.\n- output: Set export dir to this location so that the merged result can be accessed directly on the host machine.\n\n</details>\n\n### Train with Command Line Interface\n\nSee [examples/README.md](examples/README.md) for usage.\n\nUse `python src/train_bash.py -h` to display arguments description.\n\n### Deploy with OpenAI-style API and vLLM\n\n```bash\nCUDA_VISIBLE_DEVICES=0,1 API_PORT=8000 python src/api_demo.py \\\n --model_name_or_path meta-llama/Meta-Llama-3-8B-Instruct \\\n --template llama3 \\\n --infer_backend vllm \\\n --vllm_enforce_eager\n```\n\n### Download from ModelScope Hub\n\nIf you have trouble with downloading models and datasets from Hugging Face, you can use ModelScope.\n\n```bash\nexport USE_MODELSCOPE_HUB=1 # `set USE_MODELSCOPE_HUB=1` for Windows\n```\n\nTrain the model by specifying a model ID of the ModelScope Hub as the `--model_name_or_path`. You can find a full list of model IDs at [ModelScope Hub](https://modelscope.cn/models), e.g., `LLM-Research/Meta-Llama-3-8B-Instruct`.\n\n## Projects using LLaMA Factory\n\nIf you have a project that should be incorporated, please contact via email or create a pull request.\n\n<details><summary>Click to show</summary>\n\n1. Wang et al. ESRL: Efficient Sampling-based Reinforcement Learning for Sequence Generation. 2023. [[arxiv]](https://arxiv.org/abs/2308.02223)\n1. Yu et al. Open, Closed, or Small Language Models for Text Classification? 2023. [[arxiv]](https://arxiv.org/abs/2308.10092)\n1. Wang et al. UbiPhysio: Support Daily Functioning, Fitness, and Rehabilitation with Action Understanding and Feedback in Natural Language. 2023. [[arxiv]](https://arxiv.org/abs/2308.10526)\n1. Luceri et al. Leveraging Large Language Models to Detect Influence Campaigns in Social Media. 2023. [[arxiv]](https://arxiv.org/abs/2311.07816)\n1. Zhang et al. Alleviating Hallucinations of Large Language Models through Induced Hallucinations. 2023. [[arxiv]](https://arxiv.org/abs/2312.15710)\n1. Wang et al. Know Your Needs Better: Towards Structured Understanding of Marketer Demands with Analogical Reasoning Augmented LLMs. 2024. [[arxiv]](https://arxiv.org/abs/2401.04319)\n1. Wang et al. CANDLE: Iterative Conceptualization and Instantiation Distillation from Large Language Models for Commonsense Reasoning. 2024. [[arxiv]](https://arxiv.org/abs/2401.07286)\n1. Choi et al. FACT-GPT: Fact-Checking Augmentation via Claim Matching with LLMs. 2024. [[arxiv]](https://arxiv.org/abs/2402.05904)\n1. Zhang et al. AutoMathText: Autonomous Data Selection with Language Models for Mathematical Texts. 2024. [[arxiv]](https://arxiv.org/abs/2402.07625)\n1. Lyu et al. KnowTuning: Knowledge-aware Fine-tuning for Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.11176)\n1. Yang et al. LaCo: Large Language Model Pruning via Layer Collaps. 2024. [[arxiv]](https://arxiv.org/abs/2402.11187)\n1. Bhardwaj et al. Language Models are Homer Simpson! Safety Re-Alignment of Fine-tuned Language Models through Task Arithmetic. 2024. [[arxiv]](https://arxiv.org/abs/2402.11746)\n1. Yang et al. Enhancing Empathetic Response Generation by Augmenting LLMs with Small-scale Empathetic Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.11801)\n1. Yi et al. Generation Meets Verification: Accelerating Large Language Model Inference with Smart Parallel Auto-Correct Decoding. 2024. [[arxiv]](https://arxiv.org/abs/2402.11809)\n1. Cao et al. Head-wise Shareable Attention for Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.11819)\n1. Zhang et al. Enhancing Multilingual Capabilities of Large Language Models through Self-Distillation from Resource-Rich Languages. 2024. [[arxiv]](https://arxiv.org/abs/2402.12204)\n1. Kim et al. Efficient and Effective Vocabulary Expansion Towards Multilingual Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.14714)\n1. Yu et al. KIEval: A Knowledge-grounded Interactive Evaluation Framework for Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2402.15043)\n1. Huang et al. Key-Point-Driven Data Synthesis with its Enhancement on Mathematical Reasoning. 2024. [[arxiv]](https://arxiv.org/abs/2403.02333)\n1. Duan et al. Negating Negatives: Alignment without Human Positive Samples via Distributional Dispreference Optimization. 2024. [[arxiv]](https://arxiv.org/abs/2403.03419)\n1. Xie and Schwertfeger. Empowering Robotics with Large Language Models: osmAG Map Comprehension with LLMs. 2024. [[arxiv]](https://arxiv.org/abs/2403.08228)\n1. Zhang et al. EDT: Improving Large Language Models' Generation by Entropy-based Dynamic Temperature Sampling. 2024. [[arxiv]](https://arxiv.org/abs/2403.14541)\n1. Weller et al. FollowIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions. 2024. [[arxiv]](https://arxiv.org/abs/2403.15246)\n1. Hongbin Na. CBT-LLM: A Chinese Large Language Model for Cognitive Behavioral Therapy-based Mental Health Question Answering. 2024. [[arxiv]](https://arxiv.org/abs/2403.16008)\n1. Zan et al. CodeS: Natural Language to Code Repository via Multi-Layer Sketch. 2024. [[arxiv]](https://arxiv.org/abs/2403.16443)\n1. Liu et al. Extensive Self-Contrast Enables Feedback-Free Language Model Alignment. 2024. [[arxiv]](https://arxiv.org/abs/2404.00604)\n1. Luo et al. BAdam: A Memory Efficient Full Parameter Training Method for Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2404.02827)\n1. Du et al. Chinese Tiny LLM: Pretraining a Chinese-Centric Large Language Model. 2024. [[arxiv]](https://arxiv.org/abs/2404.04167)\n1. Liu et al. Dynamic Generation of Personalities with Large Language Models. 2024. [[arxiv]](https://arxiv.org/abs/2404.07084)\n1. **[StarWhisper](https://github.com/Yu-Yang-Li/StarWhisper)**: A large language model for Astronomy, based on ChatGLM2-6B and Qwen-14B.\n1. **[DISC-LawLLM](https://github.com/FudanDISC/DISC-LawLLM)**: A large language model specialized in Chinese legal domain, based on Baichuan-13B, is capable of retrieving and reasoning on legal knowledge.\n1. **[Sunsimiao](https://github.com/thomas-yanxin/Sunsimiao)**: A large language model specialized in Chinese medical domain, based on Baichuan-7B and ChatGLM-6B.\n1. **[CareGPT](https://github.com/WangRongsheng/CareGPT)**: A series of large language models for Chinese medical domain, based on LLaMA2-7B and Baichuan-13B.\n1. **[MachineMindset](https://github.com/PKU-YuanGroup/Machine-Mindset/)**: A series of MBTI Personality large language models, capable of giving any LLM 16 different personality types based on different datasets and training methods.\n\n</details>\n\n## License\n\nThis repository is licensed under the [Apache-2.0 License](LICENSE).\n\nPlease follow the model licenses to use the corresponding model weights: [Baichuan2](https://huggingface.co/baichuan-inc/Baichuan2-7B-Base/blob/main/Community%20License%20for%20Baichuan%202%20Model.pdf) / [BLOOM](https://huggingface.co/spaces/bigscience/license) / [ChatGLM3](https://github.com/THUDM/ChatGLM3/blob/main/MODEL_LICENSE) / [Command-R](https://cohere.com/c4ai-cc-by-nc-license) / [DeepSeek](https://github.com/deepseek-ai/DeepSeek-LLM/blob/main/LICENSE-MODEL) / [Falcon](https://huggingface.co/tiiuae/falcon-180B/blob/main/LICENSE.txt) / [Gemma](https://ai.google.dev/gemma/terms) / [InternLM2](https://github.com/InternLM/InternLM#license) / [LLaMA](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md) / [LLaMA-2/LLaVA-1.5](https://ai.meta.com/llama/license/) / [LLaMA-3](https://llama.meta.com/llama3/license/) / [Mistral](LICENSE) / [OLMo](LICENSE) / [Phi-1.5/2](https://huggingface.co/microsoft/phi-1_5/resolve/main/Research%20License.docx) / [Phi-3](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct/blob/main/LICENSE) / [Qwen](https://github.com/QwenLM/Qwen/blob/main/Tongyi%20Qianwen%20LICENSE%20AGREEMENT) / [StarCoder2](https://huggingface.co/spaces/bigcode/bigcode-model-license-agreement) / [XVERSE](https://github.com/xverse-ai/XVERSE-13B/blob/main/MODEL_LICENSE.pdf) / [Yi](https://huggingface.co/01-ai/Yi-6B/blob/main/LICENSE) / [Yuan](https://github.com/IEIT-Yuan/Yuan-2.0/blob/main/LICENSE-Yuan)\n\n## Citation\n\nIf this work is helpful, please kindly cite as:\n\n```bibtex\n@article{zheng2024llamafactory,\n title={LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models},\n author={Yaowei Zheng and Richong Zhang and Junhao Zhang and Yanhan Ye and Zheyan Luo and Yongqiang Ma},\n journal={arXiv preprint arXiv:2403.13372},\n year={2024},\n url={http://arxiv.org/abs/2403.13372}\n}\n```\n\n## Acknowledgement\n\nThis repo benefits from [PEFT](https://github.com/huggingface/peft), [TRL](https://github.com/huggingface/trl), [QLoRA](https://github.com/artidoro/qlora) and [FastChat](https://github.com/lm-sys/FastChat). Thanks for their wonderful works.\n\n## Star History\n\n\n",
"bugtrack_url": null,
"license": "Apache 2.0 License",
"summary": "Easy-to-use LLM fine-tuning framework",
"version": "0.7.0",
"project_urls": {
"Homepage": "https://github.com/hiyouga/LLaMA-Factory"
},
"split_keywords": [
"llama",
" bloom",
" falcon",
" llm",
" chatgpt",
" transformer",
" pytorch",
" deep learning"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "14f5a9bba99387032b7a929067f33ae249dd8aa005b499cdf1ff7ea83a74b1b4",
"md5": "75de06a34023769c559d00de5b870693",
"sha256": "a674ec73beb01c00e492366224064d742950d1a34717cb7f044a15b1a81f9c49"
},
"downloads": -1,
"filename": "llmtuner-0.7.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "75de06a34023769c559d00de5b870693",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8.0",
"size": 156001,
"upload_time": "2024-04-27T19:50:37",
"upload_time_iso_8601": "2024-04-27T19:50:37.371632Z",
"url": "https://files.pythonhosted.org/packages/14/f5/a9bba99387032b7a929067f33ae249dd8aa005b499cdf1ff7ea83a74b1b4/llmtuner-0.7.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "e009140b915097765b07e8e3e01ca9c433722892be4d0437e91803f1b2de848c",
"md5": "5ee30da6964120858b21d7293c44a485",
"sha256": "b30b9f9c93318d00e2d560bdf0e12d534f13c33bc1a53804c9333e6626023bf4"
},
"downloads": -1,
"filename": "llmtuner-0.7.0.tar.gz",
"has_sig": false,
"md5_digest": "5ee30da6964120858b21d7293c44a485",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8.0",
"size": 144240,
"upload_time": "2024-04-27T19:50:40",
"upload_time_iso_8601": "2024-04-27T19:50:40.169815Z",
"url": "https://files.pythonhosted.org/packages/e0/09/140b915097765b07e8e3e01ca9c433722892be4d0437e91803f1b2de848c/llmtuner-0.7.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-04-27 19:50:40",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "hiyouga",
"github_project": "LLaMA-Factory",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"requirements": [
{
"name": "torch",
"specs": [
[
">=",
"1.13.1"
]
]
},
{
"name": "transformers",
"specs": [
[
">=",
"4.37.2"
]
]
},
{
"name": "datasets",
"specs": [
[
">=",
"2.14.3"
]
]
},
{
"name": "accelerate",
"specs": [
[
">=",
"0.27.2"
]
]
},
{
"name": "peft",
"specs": [
[
">=",
"0.10.0"
]
]
},
{
"name": "trl",
"specs": [
[
">=",
"0.8.1"
]
]
},
{
"name": "gradio",
"specs": [
[
">=",
"4.0.0"
]
]
},
{
"name": "scipy",
"specs": []
},
{
"name": "einops",
"specs": []
},
{
"name": "sentencepiece",
"specs": []
},
{
"name": "protobuf",
"specs": []
},
{
"name": "uvicorn",
"specs": []
},
{
"name": "pydantic",
"specs": []
},
{
"name": "fastapi",
"specs": []
},
{
"name": "sse-starlette",
"specs": []
},
{
"name": "matplotlib",
"specs": []
},
{
"name": "fire",
"specs": []
},
{
"name": "packaging",
"specs": []
}
],
"lcname": "llmtuner"
}
JSON
