ChatDocs

Chat with your documents offline using AI. No data leaves your system. Internet connection is only required to install the tool and download the AI models.

It is based on PrivateGPT but uses C Transformers which supports more GGML models. It also supports 🤗 Transformers.

Installation

Install the tool using:

pip install chatdocs

Download the AI models using:

chatdocs download

Now it can be run offline without internet connection.

Basic Usage

Add a directory containing documents to chat with using:

chatdocs add /path/to/documents

Chat with your documents using:

chatdocs chat

Advanced Usage

To see available options for each command, run:

chatdocs add --help
chatdocs chat --help

LLM

By default it uses the Wizard-Vicuna-7B-Uncensored-GGML model which can be changed using the --model option:

chatdocs chat --model TheBloke/MPT-7B-GGML --download

Note: The --download option is only required when you use a model for the first time.

It can also be used with an existing local model file:

chatdocs chat --model /path/to/mpt-7b-ggml.bin --model-type mpt

LLM Provider

By default it uses C Transformers as the LLM provider which can be changed to 🤗 Transformers using the --hf option:

chatdocs chat --hf --model TheBloke/Wizard-Vicuna-7B-Uncensored-HF --download

DB

By default it stores the processed documents in db directory which can be changed using the --db option:

chatdocs add documents --db /path/to/some/directory
chatdocs chat --db /path/to/some/directory

UI

Coming Soon

License

MIT