> For the complete documentation index, see [llms.txt](https://docs.datasaur.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.datasaur.ai/compatibility-and-updates/release-notes/version-6/6.92.0.md).

# 6.92.0

December 12, 2024

## **What's new** 💡

### **Data Studio**

* Bounding Box Labeling — Users can now [hide individual bounding boxes](https://github.com/vulcan-ai/gitbook/blob/base/compatibility-and-updates/nlp-projects/lets-get-labeling/bounding-box-labeling.md#how-to-hide-bounding-boxes), making it easier to edit overlapping annotations during labeling tasks.
* Bounding box label set dialog in project — Enhanced the appearance of the label set dialog for configuring new or existing label set within a project.
* Document and Row Labeling Extension — Improved spacing between the question field and the bottom of the extension after project completion.
* Self-Assignment — Added the ability to [self-unassign](https://github.com/vulcan-ai/gitbook/blob/base/compatibility-and-updates/workspace-management/project-management/self-assignment/self-unassign.md) incomplete documents from the Projects page.
* Sign-In Page — Enhanced styling for the Google button when disabled.

### **LLM Labs**

* Direct Access LLMs — Integrated the new Amazon Nova model into the Amazon Bedrock provider.
* Direct Access LLMs — Integrated the new Llama 3.3 model into the Hugging Face provider.
* Evaluation — Added a new "Use in Sandbox" button, enabling users to easily redirect back to Sandbox and improve evaluation results.
* Evaluation — Improved the evaluation details dialog to display performance for each individual row.
* Knowledge base — Added OCR (Optical Character Recognition) capabilities to the knowledge base, allowing users to utilize scanned images for the RAG (Retrieval-Augmented Generation) process in LLM Labs.
* Knowledge base — Enhanced the progress loading icon in the knowledge base to align with the delete icon.
* Manual evaluation — Improved the user flow for the manual evaluation feature.
* Models catalog — Improved the Models catalog section by removing excessive empty space.
* Sandbox — Added a new "Reset to default" button in the application configuration.
* Sandbox — Added capability to support mathematical symbols and equations in completion results.

## **Bug fixes** 🐞

### **Data Studio**

* Fixed an issue where document count didn’t update in real-time during self-assignment.
* Fixed an issue where number inputs edited via keyboard weren’t correctly submitted using the Enter key in self-assignment dialog.

### **LLM Labs**

* Fixed an issue where model cards disappeared when users refreshed the page.
* Fixed an issue where the evaluation was deleted when the deployed sandbox was deleted.
* Fixed an issue where the performance monitoring project link was incorrect.
* Fixed an issue where users could not delete multiple members in LLM Labs.