DocuChat Logo

Changelog - v3.2.0

Improved sorting, Google Drive integration, analytics, and document management. Introduced new embedding model and chunking method.
Last edited: 10/11/2024
We believe in creating a tool that speaks for itself, powered and shaped by your support and feedback. So please don't hesitate to reach out to us with your thoughts and suggestions.

āœØ New

  • Advanced Embedding Model: We've upgraded to the state-of-the-art Cohere Embed Multilingual v3 model, hosted on our AWS servers in Frankfurt. This change is expected to significantly improve answer quality. All existing documents have been re-embedded with this model at no additional cost.
  • Proprietary Chunking Method: We've introduced a new chunking strategy to enhance retrieval capacity. While existing documents retain their current chunks, all newly added documents will benefit from this improved method.
  • Account Deletion for Super Admins: Super admins can now directly delete their accounts and associated information from their account settings, without needing to contact support.

šŸ› ļø Improved

  • Table Sorting Logic: Fixed a bug in our table sorting mechanism. Sorting now applies to all items, not just the current page, with pagination applied afterwards.
  • Google Drive Integration:
    • Sync failure reasons are now displayed in the Google Drive document details modal.
    • Large files that previously caused errors can now be synced successfully.
  • Team and Plan Pages: Added seat limit information for better resource management.
  • Analytics Enhancements: In the Sessions view, chatbot messages now display the sources used for answers, similar to the citation feature in normal chat.
  • Document Management:
    • Changed the default row click action in the documents page to "rename document" for improved intuitiveness.
    • Implemented various UI improvements in documents and chatbots pages for clearer actions.

šŸ› Fixed

  • Fixed several bugs and issues to enhance overall system stability and performance.