We are excited to announce the release of DocsGPT v2.0, which brings new features and improvements, including a built-in file ingester, Docker support, and ingester changes. This update also addresses bug fixes and stability improvements, making the user experience better than ever.
DocsGPT v2.0 now features a built-in file ingester, making it easier to work with documentation files. The new UI supports pdf, rst, md, and txt formats.Check out the live version
Setting up your self-hosted DocsGPT instance is now simpler than ever, thanks to Docker support. Run DocsGPT in just three steps:
Download the repository with
git clone https://github.com/arc53/DocsGPT.git
Open docker-compose.yaml and replace
your_api_key with your OpenAI's key (there are 4 instances)
docker-compose build && docker-compose up
The latest update includes new ingester options that allow for better handling of document tokens:
--token-check: Whether to group small documents and split large ones, improving semantics
--min_tokens: Minimum number of tokens to not group
--max_tokens: Maximum number of tokens to not split
DocsGPT v2.0 also includes bug fixes, stability improvements, and a migration of the Chrome extension from Manifest V2 to V3.
We would like to thank the community for their continued support and active participation. Special thanks go to our contributors, whose hard work has made these updates possible. We encourage you to try out DocsGPT v2.0 and experience the improvements for yourself!