search

lm-sys / FastChat

An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.
Apache License 2.0
37.03k stars 4.56k forks source link

readme

FastChat

| Demo | Discord | X |

FastChat is an open platform for training, serving, and evaluating large language model based chatbots.

FastChat's core features include:

News

More - [2023/08] We released **LongChat v1.5** based on Llama 2 with 32K context lengths. Download [weights](#longchat). - [2023/06] We introduced **MT-bench**, a challenging multi-turn question set for evaluating chatbots. Check out the blog [post](https://lmsys.org/blog/2023-06-22-leaderboard/). - [2023/06] We introduced **LongChat**, our long-context chatbots and evaluation tools. Check out the blog [post](https://lmsys.org/blog/2023-06-29-longchat/). - [2023/05] We introduced **Chatbot Arena** for battles among LLMs. Check out the blog [post](https://lmsys.org/blog/2023-05-03-arena). - [2023/03] We released **Vicuna: An Open-Source Chatbot Impressing GPT-4 with 90% ChatGPT Quality**. Check out the blog [post](https://vicuna.lmsys.org).

Contents

Install

Method 1: With pip

pip3 install "fschat[model_worker,webui]"

Method 2: From source

  1. Clone this repository and navigate to the FastChat folder 
    git clone https://github.com/lm-sys/FastChat.git
cd FastChat

If you are running on Mac:

brew install rust cmake
  1. Install Package 
    pip3 install --upgrade pip  # enable PEP 660 support
pip3 install -e ".[model_worker,webui]"

Model Weights

Vicuna Weights

Vicuna is based on Llama 2 and should be used under Llama's model license.

You can use the commands below to start chatting. It will automatically download the weights from Hugging Face repos. Downloaded weights are stored in a .cache folder in the user's home folder (e.g., ~/.cache/huggingface/hub/<model_name>).

See more command options and how to handle out-of-memory in the "Inference with Command Line Interface" section below.

NOTE: transformers>=4.31 is required for 16K versions.

Size Chat Command Hugging Face Repo
7B python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5 lmsys/vicuna-7b-v1.5
7B-16k python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5-16k lmsys/vicuna-7b-v1.5-16k
13B python3 -m fastchat.serve.cli --model-path lmsys/vicuna-13b-v1.5 lmsys/vicuna-13b-v1.5
13B-16k python3 -m fastchat.serve.cli --model-path lmsys/vicuna-13b-v1.5-16k lmsys/vicuna-13b-v1.5-16k
33B python3 -m fastchat.serve.cli --model-path lmsys/vicuna-33b-v1.3 lmsys/vicuna-33b-v1.3

Old weights: see docs/vicuna_weights_version.md for all versions of weights and their differences.

Other Models

Besides Vicuna, we also released two additional models: LongChat and FastChat-T5. You can use the commands below to chat with them. They will automatically download the weights from Hugging Face repos.

Model Chat Command Hugging Face Repo
LongChat-7B python3 -m fastchat.serve.cli --model-path lmsys/longchat-7b-32k-v1.5 lmsys/longchat-7b-32k
FastChat-T5-3B python3 -m fastchat.serve.cli --model-path lmsys/fastchat-t5-3b-v1.0 lmsys/fastchat-t5-3b-v1.0

Inference with Command Line Interface

(Experimental Feature: You can specify --style rich to enable rich text output and better text streaming quality for some non-ASCII content. This may not work properly on certain terminals.)

Supported Models

FastChat supports a wide range of models, including LLama 2, Vicuna, Alpaca, Baize, ChatGLM, Dolly, Falcon, FastChat-T5, GPT4ALL, Guanaco, MTP, OpenAssistant, OpenChat, RedPajama, StableLM, WizardLM, xDAN-AI and more.

See a complete list of supported models and instructions to add a new model here.

Single GPU

The command below requires around 14GB of GPU memory for Vicuna-7B and 28GB of GPU memory for Vicuna-13B. See the "Not Enough Memory" section below if you do not have enough memory. --model-path can be a local folder or a Hugging Face repo name.

python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5

Multiple GPUs

You can use model parallelism to aggregate GPU memory from multiple GPUs on the same machine. 

python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5 --num-gpus 2

Tips: Sometimes the "auto" device mapping strategy in huggingface/transformers does not perfectly balance the memory allocation across multiple GPUs. You can use --max-gpu-memory to specify the maximum memory per GPU for storing model weights. This allows it to allocate more memory for activations, so you can use longer context lengths or larger batch sizes. For example,

python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5 --num-gpus 2 --max-gpu-memory 8GiB

CPU Only

This runs on the CPU only and does not require GPU. It requires around 30GB of CPU memory for Vicuna-7B and around 60GB of CPU memory for Vicuna-13B.

python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5 --device cpu

Use Intel AI Accelerator AVX512_BF16/AMX to accelerate CPU inference.

CPU_ISA=amx python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5 --device cpu

Metal Backend (Mac Computers with Apple Silicon or AMD GPUs)

Use --device mps to enable GPU acceleration on Mac computers (requires torch >= 2.0). Use --load-8bit to turn on 8-bit compression.

python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5 --device mps --load-8bit

Vicuna-7B can run on a 32GB M1 Macbook with 1 - 2 words / second.

Intel XPU (Intel Data Center and Arc A-Series GPUs)

Install the Intel Extension for PyTorch. Set the OneAPI environment variables:

source /opt/intel/oneapi/setvars.sh

Use --device xpu to enable XPU/GPU acceleration.

python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5 --device xpu

Vicuna-7B can run on an Intel Arc A770 16GB.

Ascend NPU

Install the Ascend PyTorch Adapter. Set the CANN environment variables:

source /usr/local/Ascend/ascend-toolkit/set_env.sh

Use --device npu to enable NPU acceleration.

python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5 --device npu

Vicuna-7B/13B can run on an Ascend NPU.

Not Enough Memory

If you do not have enough memory, you can enable 8-bit compression by adding --load-8bit to commands above. This can reduce memory usage by around half with slightly degraded model quality. It is compatible with the CPU, GPU, and Metal backend.

Vicuna-13B with 8-bit compression can run on a single GPU with 16 GB of VRAM, like an Nvidia RTX 3090, RTX 4080, T4, V100 (16GB), or an AMD RX 6800 XT.

python3 -m fastchat.serve.cli --model-path lmsys/vicuna-7b-v1.5 --load-8bit

In addition to that, you can add --cpu-offloading to commands above to offload weights that don't fit on your GPU onto the CPU memory. This requires 8-bit compression to be enabled and the bitsandbytes package to be installed, which is only available on linux operating systems.

More Platforms and Quantization

Use models from modelscope

For Chinese users, you can use models from www.modelscope.cn via specify the following environment variables.

export FASTCHAT_USE_MODELSCOPE=True

Serving with Web GUI

To serve using the web UI, you need three main components: web servers that interface with users, model workers that host one or more models, and a controller to coordinate the webserver and model workers. You can learn more about the architecture here.

Here are the commands to follow in your terminal:

Launch the controller

python3 -m fastchat.serve.controller

This controller manages the distributed workers.

Launch the model worker(s)

python3 -m fastchat.serve.model_worker --model-path lmsys/vicuna-7b-v1.5

Wait until the process finishes loading the model and you see "Uvicorn running on ...". The model worker will register itself to the controller .

To ensure that your model worker is connected to your controller properly, send a test message using the following command:

python3 -m fastchat.serve.test_message --model-name vicuna-7b-v1.5

You will see a short output.

Launch the Gradio web server

python3 -m fastchat.serve.gradio_web_server

This is the user interface that users will interact with.

By following these steps, you will be able to serve your models using the web UI. You can open your browser and chat with a model now. If the models do not show up, try to reboot the gradio web server.

Launch Chatbot Arena (side-by-side battle UI)

Currently, Chatbot Arena is powered by FastChat. Here is how you can launch an instance of Chatbot Arena locally.

FastChat supports popular API-based models such as OpenAI, Anthropic, Gemini, Mistral and more. To add a custom API, please refer to the model support doc. Below we take OpenAI models as an example.

Create a JSON configuration file api_endpoint.json with the api endpoints of the models you want to serve, for example:

{
    "gpt-4o-2024-05-13": {
        "model_name": "gpt-4o-2024-05-13",
        "api_base": "https://api.openai.com/v1",
        "api_type": "openai",
        "api_key": [Insert API Key],
        "anony_only": false
    }
}

For Anthropic models, specify "api_type": "anthropic_message" with your Anthropic key. Similarly, for gemini model, specify "api_type": "gemini". More details can be found in api_provider.py.

To serve your own model using local gpus, follow the instructions in Serving with Web GUI.

Now you're ready to launch the server:

python3 -m fastchat.serve.gradio_web_server_multi --register-api-endpoint-file api_endpoint.json

(Optional): Advanced Features, Scalability, Third Party UI

API

OpenAI-Compatible RESTful APIs & SDK

FastChat provides OpenAI-compatible APIs for its supported models, so you can use FastChat as a local drop-in replacement for OpenAI APIs. The FastChat server is compatible with both openai-python library and cURL commands. The REST API is capable of being executed from Google Colab free tier, as demonstrated in the FastChat_API_GoogleColab.ipynb notebook, available in our repository. See docs/openai_api.md.

Hugging Face Generation APIs

See fastchat/serve/huggingface_api.py.

LangChain Integration

See docs/langchain_integration.

Evaluation

We use MT-bench, a set of challenging multi-turn open-ended questions to evaluate models. To automate the evaluation process, we prompt strong LLMs like GPT-4 to act as judges and assess the quality of the models' responses. See instructions for running MT-bench at fastchat/llm_judge.

MT-bench is the new recommended way to benchmark your models. If you are still looking for the old 80 questions used in the vicuna blog post, please go to vicuna-blog-eval.

Fine-tuning

Data

Vicuna is created by fine-tuning a Llama base model using approximately 125K user-shared conversations gathered from ShareGPT.com with public APIs. To ensure data quality, we convert the HTML back to markdown and filter out some inappropriate or low-quality samples. Additionally, we divide lengthy conversations into smaller segments that fit the model's maximum context length. For detailed instructions to clean the ShareGPT data, check out here.

We will not release the ShareGPT dataset. If you would like to try the fine-tuning code, you can run it with some dummy conversations in dummy_conversation.json. You can follow the same format and plug in your own data.

Code and Hyperparameters

Our code is based on Stanford Alpaca with additional support for multi-turn conversations. We use similar hyperparameters as the Stanford Alpaca.

Hyperparameter Global Batch Size Learning rate Epochs Max length Weight decay
Vicuna-13B 128 2e-5 3 2048 0

Fine-tuning Vicuna-7B with Local GPUs

Tips:

Other models, platforms and LoRA support

More instructions to train other models (e.g., FastChat-T5) and use LoRA are in docs/training.md.

Fine-tuning on Any Cloud with SkyPilot

SkyPilot is a framework built by UC Berkeley for easily and cost effectively running ML workloads on any cloud (AWS, GCP, Azure, Lambda, etc.). Find SkyPilot documentation here on using managed spot instances to train Vicuna and save on your cloud costs.

Citation

The code (training, serving, and evaluation) in this repository is mostly developed for or derived from the paper below. Please cite it if you find the repository helpful.

@misc{zheng2023judging,
      title={Judging LLM-as-a-judge with MT-Bench and Chatbot Arena},
      author={Lianmin Zheng and Wei-Lin Chiang and Ying Sheng and Siyuan Zhuang and Zhanghao Wu and Yonghao Zhuang and Zi Lin and Zhuohan Li and Dacheng Li and Eric. P Xing and Hao Zhang and Joseph E. Gonzalez and Ion Stoica},
      year={2023},
      eprint={2306.05685},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

We are also planning to add more of our research to this repository.