--- license: apache-2.0 library_name: transformers pipeline_tag: text-generation --- ![# LLaMA Factory](assets/logo.png) This repository contains the code for fine-tuning models, as described in [Autonomous Data Selection with Language Models for Mathematical Texts](https://huggingface.co/papers/2402.07625). [![GitHub Repo stars](https://img.shields.io/github/stars/hiyouga/LLaMA-Factory?style=social)](https://github.com/hiyouga/LLaMA-Factory/stargazers) [![GitHub Code License](https://img.shields.io/github/license/hiyouga/LLaMA-Factory)](LICENSE) [![GitHub last commit](https://img.shields.io/github/last-commit/hiyouga/LLaMA-Factory)](https://github.com/hiyouga/LLaMA-Factory/commits/main) [![PyPI](https://img.shields.io/pypi/v/llamafactory)](https://pypi.org/project/llamafactory/) [![Citation](https://img.shields.io/badge/citation-72-green)](#projects-using-llama-factory) [![GitHub pull request](https://img.shields.io/badge/PRs-welcome-blue)](https://github.com/hiyouga/LLaMA-Factory/pulls) [![Discord](https://dcbadge.vercel.app/api/server/rKfvV9r9FK?compact=true&style=flat)](https://discord.gg/rKfvV9r9FK) [![Twitter](https://img.shields.io/twitter/follow/llamafactory_ai)](https://twitter.com/llamafactory_ai) [![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1eRTPn37ltBbYsISy9Aw2NuI2Aq5CQrD9?usp=sharing) [![Open in DSW](https://gallery.pai-ml.com/assets/open-in-dsw.svg)](https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory) [![Spaces](https://img.shields.io/badge/πŸ€—-Open%20in%20Spaces-blue)](https://huggingface.co/spaces/hiyouga/LLaMA-Board) [![Studios](https://img.shields.io/badge/ModelScope-Open%20in%20Studios-blue)](https://modelscope.cn/studios/hiyouga/LLaMA-Board) [![GitHub Tread](https://trendshift.io/api/badge/repositories/4535)](https://trendshift.io/repositories/4535) πŸ‘‹ Join our [WeChat](assets/wechat.jpg) or [NPU user group](assets/wechat_npu.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 - **PAI-DSW**: https://gallery.pai-ml.com/#/preview/deepLearning/nlp/llama_factory - **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, KTO, ORPO, etc. - **Scalable resources**: 16-bit full-tuning, freeze-tuning, LoRA and 2/3/4/5/6/8-bit QLoRA via AQLM/AWQ/GPTQ/LLM.int8/HQQ/EETQ. - **Advanced algorithms**: GaLore, BAdam, DoRA, LongLoRA, LLaMA Pro, Mixture-of-Depths, LoRA+, LoftQ, PiSSA 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. ![benchmark](assets/benchmark.svg)
Definitions - **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.
## Changelog [24/06/16] We support **[PiSSA](https://arxiv.org/abs/2404.02948)** algorithm. See [examples](examples/README.md) for usage. [24/06/07] We supported fine-tuning the **[Qwen2](https://qwenlm.github.io/blog/qwen2/)** and **[GLM-4](https://github.com/THUDM/GLM-4)** models. [24/05/26] We supported **[SimPO](https://arxiv.org/abs/2405.14734)** algorithm for preference learning. See [examples](examples/README.md) for usage.
Full Changelog [24/05/20] We supported fine-tuning the **PaliGemma** series models. Note that the PaliGemma models are pre-trained models, you need to fine-tune them with `gemma` template for chat completion. [24/05/18] We supported **[KTO](https://arxiv.org/abs/2402.01306)** algorithm for preference learning. See [examples](examples/README.md) for usage. [24/05/14] We supported training and inference on the Ascend NPU devices. Check [installation](#installation) section for details. [24/04/26] We supported fine-tuning the **LLaVA-1.5** multimodal LLMs. See [examples](examples/README.md) 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](examples/README.md) for usage. [24/04/16] We supported **[BAdam](https://arxiv.org/abs/2404.02827)**. See [examples](examples/README.md) 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). [24/03/31] We supported **[ORPO](https://arxiv.org/abs/2403.07691)**. See [examples](examples/README.md) 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](examples/README.md) for usage. [24/03/13] We supported **[LoRA+](https://arxiv.org/abs/2402.12354)**. See [examples](examples/README.md) for usage. [24/03/07] We supported gradient low-rank projection (**[GaLore](https://arxiv.org/abs/2403.03507)**) algorithm. See [examples](examples/README.md) 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. [24/02/28] We supported weight-decomposed LoRA (**[DoRA](https://arxiv.org/abs/2402.09353)**). Try `use_dora: true` to activate DoRA training. [24/02/15] We supported **block expansion** proposed by [LLaMA Pro](https://github.com/TencentARC/LLaMA-Pro). See [examples](examples/README.md) 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_en`. [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: true` 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)**. See [this tutorial](#download-from-modelscope-hub) for usage. [23/10/21] We supported **[NEFTune](https://arxiv.org/abs/2310.05914)** trick for fine-tuning. Try `neftune_noise_alpha: 5` argument to activate NEFTune. [23/09/27] We supported **$S^2$-Attn** proposed by [LongLoRA](https://github.com/dvlab-research/LongLoRA) for the LLaMA models. Try `shift_attn: true` argument to enable shift short attention. [23/09/23] We integrated MMLU, C-Eval and CMMLU benchmarks in this repo. See [examples](examples/README.md) for usage. [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 [examples](examples/README.md) for usage. [23/07/31] We supported **dataset streaming**. Try `streaming: true` 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)** ⚑ 🩹, 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/create) 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)**). See [examples](examples/README.md) for usage.
## Supported Models | Model | Model size | Template | | ------------------------------------------------------------ | -------------------------------- | --------- | | [Baichuan 2](https://huggingface.co/baichuan-inc) | 7B/13B | baichuan2 | | [BLOOM/BLOOMZ](https://huggingface.co/bigscience) | 560M/1.1B/1.7B/3B/7.1B/176B | - | | [ChatGLM3](https://huggingface.co/THUDM) | 6B | chatglm3 | | [Command R](https://huggingface.co/CohereForAI) | 35B/104B | cohere | | [DeepSeek (Code/MoE)](https://huggingface.co/deepseek-ai) | 7B/16B/67B/236B | deepseek | | [Falcon](https://huggingface.co/tiiuae) | 7B/11B/40B/180B | falcon | | [Gemma/Gemma 2/CodeGemma](https://huggingface.co/google) | 2B/7B/9B/27B | gemma | | [GLM-4](https://huggingface.co/THUDM) | 9B | glm4 | | [InternLM2](https://huggingface.co/internlm) | 7B/20B | intern2 | | [Llama](https://github.com/facebookresearch/llama) | 7B/13B/33B/65B | - | | [Llama 2](https://huggingface.co/meta-llama) | 7B/13B/70B | llama2 | | [Llama 3](https://huggingface.co/meta-llama) | 8B/70B | llama3 | | [LLaVA-1.5](https://huggingface.co/llava-hf) | 7B/13B | vicuna | | [Mistral/Mixtral](https://huggingface.co/mistralai) | 7B/8x7B/8x22B | mistral | | [OLMo](https://huggingface.co/allenai) | 1B/7B | - | | [PaliGemma](https://huggingface.co/google) | 3B | gemma | | [Phi-1.5/Phi-2](https://huggingface.co/microsoft) | 1.3B/2.7B | - | | [Phi-3](https://huggingface.co/microsoft) | 4B/7B/14B | phi | | [Qwen/Qwen1.5/Qwen2 (Code/MoE)](https://huggingface.co/Qwen) | 0.5B/1.5B/4B/7B/14B/32B/72B/110B | qwen | | [StarCoder 2](https://huggingface.co/bigcode) | 3B/7B/15B | - | | [XVERSE](https://huggingface.co/xverse) | 7B/13B/65B | xverse | | [Yi/Yi-1.5](https://huggingface.co/01-ai) | 6B/9B/34B | yi | | [Yi-VL](https://huggingface.co/01-ai) | 6B/34B | yi_vl | | [Yuan 2](https://huggingface.co/IEITYuan) | 2B/51B/102B | yuan | > [!NOTE] > 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/llamafactory/extras/constants.py) for a full list of models we supported. You also can add a custom chat template to [template.py](src/llamafactory/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: | | KTO 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: | | SimPO Training | :white_check_mark: | :white_check_mark: | :white_check_mark: | :white_check_mark: | ## Provided Datasets
Pre-training datasets - [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) - [FineWeb (en)](https://huggingface.co/datasets/HuggingFaceFW/fineweb) - [FineWeb-Edu (en)](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu) - [The Stack (en)](https://huggingface.co/datasets/bigcode/the-stack) - [StarCoder (en)](https://huggingface.co/datasets/bigcode/starcoderdata)
Supervised fine-tuning datasets - [Identity (en&zh)](data/identity.json) - [Stanford Alpaca (en)](https://github.com/tatsu-lab/stanford_alpaca) - [Stanford Alpaca (zh)](https://github.com/ymcui/Chinese-LLaMA-Alpaca-3) - [Alpaca GPT4 (en&zh)](https://github.com/Instruction-Tuning-with-GPT-4/GPT-4-LLM) - [Glaive Function Calling V2 (en&zh)](https://huggingface.co/datasets/glaiveai/glaive-function-calling-v2) - [LIMA (en)](https://huggingface.co/datasets/GAIR/lima) - [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) - [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) - [Advertise Generating (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) - [Cosmopedia (en)](https://huggingface.co/datasets/HuggingFaceTB/cosmopedia) - [STEM (zh)](https://huggingface.co/datasets/hfl/stem_zh_instruction) - [Ruozhiba (zh)](https://huggingface.co/datasets/hfl/ruozhiba_gpt4_turbo) - [Neo-sft (zh)](https://huggingface.co/datasets/m-a-p/neo_sft_phase2) - [Magpie-Pro-300K-Filtered (en)](https://huggingface.co/datasets/Magpie-Align/Magpie-Pro-300K-Filtered) - [WebInstructSub (en)](https://huggingface.co/datasets/TIGER-Lab/WebInstructSub)
Preference datasets - [DPO mixed (en&zh)](https://huggingface.co/datasets/hiyouga/DPO-En-Zh-20k) - [UltraFeedback (en)](https://huggingface.co/datasets/HuggingFaceH4/ultrafeedback_binarized) - [Orca DPO Pairs (en)](https://huggingface.co/datasets/Intel/orca_dpo_pairs) - [HH-RLHF (en)](https://huggingface.co/datasets/Anthropic/hh-rlhf) - [Nectar (en)](https://huggingface.co/datasets/berkeley-nest/Nectar) - [Orca DPO (de)](https://huggingface.co/datasets/mayflowergmbh/intel_orca_dpo_pairs_de) - [KTO mixed (en)](https://huggingface.co/datasets/argilla/kto-mix-15k)
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 | 3.11 | | torch | 1.13.1 | 2.3.0 | | transformers | 4.41.2 | 4.41.2 | | datasets | 2.16.0 | 2.19.2 | | accelerate | 0.30.1 | 0.30.1 | | peft | 0.11.1 | 0.11.1 | | trl | 0.8.6 | 0.9.4 | | Optional | Minimum | Recommend | | ------------ | ------- | --------- | | CUDA | 11.6 | 12.2 | | deepspeed | 0.10.0 | 0.14.0 | | bitsandbytes | 0.39.0 | 0.43.1 | | vllm | 0.4.3 | 0.4.3 | | flash-attn | 2.3.0 | 2.5.9 | ### 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 |