2022-12-21 05:17:38 +01:00
# text-generation-webui
2022-12-21 20:49:30 +01:00
A gradio webui for running large language models locally. Supports gpt-j-6B, gpt-neox-20b, opt, galactica, and many others.
Its goal is to become the [AUTOMATIC1111/stable-diffusion-webui ](https://github.com/AUTOMATIC1111/stable-diffusion-webui ) of text generation.
2022-12-21 17:17:06 +01:00
2022-12-21 18:04:51 +01:00
![webui screenshot ](https://github.com/oobabooga/text-generation-webui/raw/main/webui.png )
2023-01-07 22:40:51 +01:00
## Features
* Switch between different models using a dropdown menu.
2023-01-09 20:19:57 +01:00
* Generate nice HTML output for GPT-4chan.
2023-01-07 23:19:55 +01:00
* Generate Markdown output for [GALACTICA ](https://github.com/paperswithcode/galai ), including LaTeX support.
2023-01-07 22:40:51 +01:00
* Notebook mode that resembles OpenAI's playground.
2023-01-08 02:52:46 +01:00
* Chat mode for conversation and role playing.
2023-01-07 22:40:51 +01:00
* Load 13b/20b models in 8-bit mode.
* Load parameter presets from text files.
2023-01-09 15:11:05 +01:00
* CPU mode.
2023-01-07 22:40:51 +01:00
2022-12-21 17:17:06 +01:00
## Installation
2023-01-06 04:05:11 +01:00
Create a conda environment:
conda create -n textgen
conda activate textgen
Install the appropriate pytorch for your GPU. For NVIDIA GPUs, this should work:
conda install pytorch torchvision torchaudio pytorch-cuda=11.7 -c pytorch -c nvidia
Install the requirements:
pip install -r requirements.txt
2022-12-21 17:17:06 +01:00
2022-12-21 18:37:50 +01:00
## Downloading models
2023-01-06 06:12:38 +01:00
Models should be placed under `models/model-name` . For instance, `models/gpt-j-6B` for [gpt-j-6B ](https://huggingface.co/EleutherAI/gpt-j-6B/tree/main ).
2022-12-21 18:37:50 +01:00
2022-12-21 19:00:16 +01:00
#### Hugging Face
2022-12-21 18:37:50 +01:00
2023-01-06 06:13:48 +01:00
Hugging Face is the main place to download models. These are some of my favorite:
2022-12-21 18:37:50 +01:00
2023-01-06 04:13:26 +01:00
* [gpt-j-6B ](https://huggingface.co/EleutherAI/gpt-j-6B/tree/main )
* [gpt-neox-20b ](https://huggingface.co/EleutherAI/gpt-neox-20b/tree/main )
2023-01-07 23:19:55 +01:00
* [OPT ](https://huggingface.co/models?search=facebook/opt )
* [GALACTICA ](https://huggingface.co/models?search=facebook/galactica )
2023-01-06 04:13:26 +01:00
* [\*-Erebus ](https://huggingface.co/models?search=erebus )
2023-01-06 06:12:38 +01:00
The files that you need to download are the json, txt, and pytorch\*.bin files. The remaining files are not necessary.
2022-12-21 18:37:50 +01:00
2023-01-07 00:06:24 +01:00
For your convenience, you can automatically download a model from HF using the script `download-model.py` . Its usage is very simple:
2023-01-07 00:00:58 +01:00
python download-model.py organization/model
For instance:
python download-model.py facebook/opt-1.3b
2023-01-09 20:19:57 +01:00
#### GPT-4chan
2022-12-21 18:37:50 +01:00
2023-01-09 20:19:57 +01:00
[GPT-4chan ](https://huggingface.co/ykilcher/gpt-4chan ) has been shut down from Hugging Face, so you need to download it elsewhere. You have two options:
2022-12-21 18:37:50 +01:00
* Torrent: [16-bit ](https://archive.org/details/gpt4chan_model_float16 ) / [32-bit ](https://archive.org/details/gpt4chan_model )
* Direct download: [16-bit ](https://theswissbay.ch/pdf/_notpdf_/gpt4chan_model_float16/ ) / [32-bit ](https://theswissbay.ch/pdf/_notpdf_/gpt4chan_model/ )
2023-01-08 03:13:43 +01:00
You also need to put GPT-J-6B's config.json file in the same folder: [config.json ](https://huggingface.co/EleutherAI/gpt-j-6B/raw/main/config.json )
2023-01-09 22:58:24 +01:00
#### Converting to pytorch (optional)
2022-12-21 20:49:30 +01:00
2023-01-06 05:41:52 +01:00
The script `convert-to-torch.py` allows you to convert models to .pt format, which is about 10x faster to load:
2022-12-21 20:49:30 +01:00
2023-01-08 18:37:43 +01:00
python convert-to-torch.py models/model-name
2022-12-21 20:49:30 +01:00
2023-01-08 18:37:43 +01:00
The output model will be saved to `torch-dumps/model-name.pt` . When you load a new model, the webui first looks for this .pt file; if it is not found, it loads the model as usual from `models/model-name` .
2022-12-21 20:49:30 +01:00
2022-12-21 17:17:06 +01:00
## Starting the webui
conda activate textgen
python server.py
2023-01-07 00:00:58 +01:00
Then browse to
`http://localhost:7860/?__theme=dark`
2023-01-07 00:07:58 +01:00
Optionally, you can use the following command-line flags:
2023-01-07 00:00:58 +01:00
2023-01-07 00:29:00 +01:00
`--model model-name` : Load this model by default.
`--notebook` : Launch the webui in notebook mode, where the output is written to the same text box as the input.
2022-12-21 20:52:23 +01:00
2023-01-08 02:52:46 +01:00
`--chat` : Launch the webui in chat mode.
2023-01-09 14:58:46 +01:00
`--cpu` : Use the CPU to generate text instead of the GPU.
2023-01-06 05:33:21 +01:00
## Presets
Inference settings presets can be created under `presets/` as text files. These files are detected automatically at startup.
2023-01-09 22:12:41 +01:00
## System requirements
2023-01-09 22:18:35 +01:00
Check the [wiki ](https://github.com/oobabooga/text-generation-webui/wiki/System-requirements ) for some examples of VRAM and RAM usage in both GPU and CPU mode.
2023-01-09 22:12:41 +01:00
2022-12-21 20:52:23 +01:00
## Contributing
2023-01-09 20:09:27 +01:00
Pull requests, suggestions and issue reports are welcome.
2023-01-09 20:15:54 +01:00
## Other projects
Make sure to also check out the great work by [KoboldAI ](https://github.com/KoboldAI/KoboldAI-Client ). I have borrowed some of the presets listed on their [wiki ](https://github.com/KoboldAI/KoboldAI-Client/wiki/Settings-Presets ) after performing a k-means clustering analysis to select the most relevant subsample.