​
April 9, 2025

​
Feature Updates

  • Now have code editor view for README and when Python Notebook has syntax errors
Editing README Generating images in Model Inference Anthropic in Model Inference
  • Added full screen option to eval DF view
Full screen in Model Inference
  • Added Support for Model Deprecation
  • Allow import of hugging face datasets into Server
  • Added a copy prompt button to entries in the eval list view
  • For Notebooks added branch indicator and now respects branches with slashes
  • Now able to merge changes on the server where there are no file conflicts (for example: a notebook is staged and a data frame is staged)
  • GitHub Actions for python windows
  • Expanded row-view component to make viewing complex content easier
  • Added ability to define code-language in field metadata
  • Created markdown and code view renderer for Row Details view
  • Added drop down to select code language in Code view renderer
  • Added support for paging through records (forward, back arrows)
  • Updated the Model Evaluation Setup form to show developer dropdown of models
  • Now showing proper status indicators on eval cells with image metadata
  • Extended Branch Picker component to support workspaces
  • Upgraded the build caching mechanism to boost performance and compatibility. #70
  • Improved file formatting for consistency.#70
  • Introduced a new capability to verify branch existence, enhancing branch management. #71
  • Standardized file path construction using modern utilities, improving clarity and maintainability. #71
  • Enhanced branch management with improved checks to prevent conflicts. #72
  • Refined file addition functionality to seamlessly support adding folders as well as individual files. #72
  • Enhanced data handling by automatically creating records when needed and improving error feedback. #73
  • Introduced more flexible branch management during data operations and workspace initialization for a smoother user experience. #73
  • Enhanced the workspace creation process, now generating unique workspace names to help avoid naming conflicts. #75
  • Introduced new properties in the Workspace class for easier access to workspace attributes: ID, name, branch, and associated repository. #75
  • Modified branch switching behavior to ensure a more consistent experience when creating and switching branches. #75
  • Streamlined access to repository and branch attributes in the DataFrame class for improved readability and performance. #75
  • Data operations now provide immediate feedback after inserting rows. #76
  • Users can specify workspace names during repository actions. #76
  • Workspace summaries now display associated commit identifiers. #76
  • Introduced the option to initialize empty directory listings. #76
  • Added methods to generate workspace URLs and to retrieve individual dataset rows. #77
  • Enhanced repository operations with improved commit tracking, merge handling, and branch management. #77
  • Introduced functionality to create empty paginated directory entries. #77
  • Expanded workspace interfaces to include commit identifier details. #77
  • Added a new error handling mechanism for data retrieval methods. #77

​
By Customer Request

  • Added how to save a parquet file to a directory, and errors with to_csv in Python by Mirascope’s request
  • Added method to test if a branch already exists
  • Create data frame if it does not exist on the first insert

​
Fixed Bugs

  • Fixed oxen pull/checkout so it doesn’t overwrite modified files
  • Fixed Model Inference committing errors
  • Edit file was not working on data frames with binary data in columns
  • Was Failing to push and pull entries that are exactly AVG_CHUNK_SIZE
  • Fixed error on using a struct field in your Model Inference prompt
  • Fixed file upload does not land in the correct place with many subdirectories
  • Fixed sample not working on branches with a slash eval/gemini
  • Fixed oxen rm -r so it doesn’t remove a directory at the root removes all directories on the remote after push
  • Model Inference commits to a sub/directory/file.parquet was not working
  • DataFrame image thumbnail was not rendering properly
  • Compare Nav Tab was not being underlined when clicked
  • Long URL’s in blog posts now wrap on mobile
  • Removed little scroll bar next to the tab bar nav and home page logos
  • Better handling for bad workspace state in workspaces page
  • Empty repos without any branches were breaking on the root repo page
  • Notebooks commit button was not showing commit form in full screen mode
  • On the site wide search dropdown users were listed with a slash and not navigating to the root of the user namespace
  • Embedded videos in the dives and entries pages were not being expanded to the proper height
  • Improved error handling when commit data is missing. #76
  • Streamlined asynchronous repository operations for a more reliable experience. #76

​
February 25, 2025

​
Feature Updates

  • We’ve improved our blog styling
Blog update
  • We’ve cleaned up the workspace list view
  • Re-enable “For Every GitHub star, an ox gets their wings”
Flying Oxen Repo
  • Created wrapper for datasets library in our Python SDK that allows you to use HF datasets library with Oxen.ai

​
By Customer Request

​
Fixed Bugs

  • We’ve fixed pip install oxenai so it doesn’t fail on Debian-based images in Docker builds
  • Full screen viewer button working in the UI
  • Blog H1 are now centered
  • Fixed saving parquet files to a directory and errors with to_csv

​
February 10, 2025

​
Feature Updates

  • oxen init, oxen-server start, and oxen help now have some getting started info on client and server initialization

​
By Customer Request

​
Fixed Bugs

  • We now have better handling on commit view in the UI

​
January 31, 2025

​
Feature Updates

  • We’ve added Deepseek models to Open.AI!

  • New Model Description Page! Click on any model in The Model Page to learn what to use it for, stats, and compare prices.

  • You can now download your SQL query results within workspaces and in the UI!

  • We cleaned up Oxen.ai’s navigation bar:)

  • The Query Data Frame API now accepts a query id for downloading a SQL queried data frame in a workspace

  • As per request (shout out Paul), we’ve support PUT to a file with a based-on revision passed in

  • Added —remote to oxen workspace list

  • Added ability to merge with the python library

  • By customer request we’ve slimmed down Python dependencies with FSSpec

​
Fixed Bugs

  • Fixed data type count for repos in the Oxen Hub
  • Squashed bug where commit view in the UI was breaking
  • oxen rm -r on an empty directory said “removed” but was not removed. Now it’s fixed!
  • Fixed oxen diff for modified text files
  • Fixed Remote Repo failure and RemoteRepo.add bug
  • Fixed bug for Oxen add on Windows on main branch
  • Fixed bug for uploading parquet files

​
January 3, 2025

​
Feature Updates

  • You can now run Nearest Neighbor Search on your embeddings! (Heres the repo)

  • We now have full page view for deeper data exploration! (Just click the expand button on the right)

  • Improved the onboarding flow for mobile (DW, I’ll charge my phone soon)

  • We’ve improved the data type tags so all jsonl, csv, and parquet files are now tagged as tabular

  • Updated oxen download command to be updated syntax

  • Added options to select tabular files or upload a new file

  • Added pytests to CI

  • We’ve added a python method to sort by input embedding

  • Added Bytez Models

  • Added a new field on evals for unit price in API

  • A repo is created when a file is uploaded

  • We moved error banner in Model Inference above the dataframe

  • We’ve added to the HTTP API docs

  • You can now ask Text2SQL queries throught the workspaces API and download results

​
Fixed Bugs

  • Shout out Mund for letting us know Model Inference crashed and wasn’t showing the commit message. Fixed! :)
  • when an ID is 0, the sidbar will not show Null but 0
  • The target branch stays the target branch when running an evaluation from the completed evaluation page (it used to use the evaluation ID as the target branch)
  • You can now have several workspaces with the same commit ID
  • Columns used to sometimes show up twice…not anymore:)

​
December 26, 2024

​
Feature Updates

  • New onboading flow to upload files

  • New starred repos section in your profile

  • You can now save inference results to a different file name

  • We now have APIs to sort your dataframes by embedding cosine similarity

  • The list of collaborators for a repo now marks the repo owners with a crown

  • There is now an “add Collaborators” Button

  • You can now see all the people who have starred your repo in the stargazer view

  • We removed “Versions” Tab from Top Nav

  • There is now a Commit button directly on the data frame

  • You now have more advanced commit options for editable data frames

  • We now list workspace API so admins can see all workspaces

  • There is now a Create Query API

  • Auto detect image file type with Rust for images that do not end in .png or .jpg

​
Fixed Bugs

  • Dataframe json content now matches in side bar and column
  • Improved the starred repo API updates
  • Refactored adding files to use new workspace API
  • Refactored DF Edit to use file scoped user workspace
  • Improved image rendering in dataframes
  • Found a better place for “add row” button in editable df
  • Fixed repo validation to allow ”.”
  • Cleaned up repo card to account for more categories
  • We now have a fallback “download” if we cannot render in the UI

​
December 03, 2024

​
Feature Updates

  • Forking repos now supported (button on top right, next to clone)
  • Now able to oxen add, oxen add schema metadata, and oxen rm in a subdirectory
  • Enabled pushing, pulling, merging of subtrees
  • Modified Evaluations to use input/output types in place of existing type

​
Fixed Bugs

  • Versions page no longer gives an error when no branches on the repo
  • Fixed some image rendering bugs in dataframe
  • Refactored oxen CLI commands for windows
  • Evaluations now handle long file names overflow in list view
  • Unknown column data type won’t break the server on column add
  • Data Frame json content now matches in side bar vs column

​
August 13, 2024

​
Feature Updates

  • CLI // Added command to easily set schema metadata render type

e.g. Instead of

oxen schemas add data.csv -c 'file' -m '{

"_oxen": {

    "render": {

        "func": "image"

        }

    }

}'

Now we have:

oxen schemas render data.csv -c 'file' -t 'image'

or…

oxen schemas add data.csv -c 'file' --render 'image’
  • CLI // unlocks branch with a oxen branch unlock command
  • Hub // Expose RepositoryCategory APIs
  • UI // Delete, edit, and add row on dataframe
  • UI // View staged DF diffs
  • UI // Add instructions to the README to clone repo, add first file, and push back

​
Fixed Bugs

  • Image name with ~ at the end are now rendering in UI
  • We fixed collaborator email button (couldn’t be clicked)
  • Hub // now renders images on branches
  • UI // filenames overflow in data frame image viewer
  • UI // Toggle column visibility in Dataframe (view & edit modes)
  • UI // Remove the left side redundant NLP query

​
June 7, 2024

​
Feature Updates

  • Hub // Added header for the HEAD commit ID of file
  • UI // cell navigation in data-frame is now bi-directional with row details pane
  • Added collaborator CTA

​
Fixed Bugs

  • UI // edit repository name was retuning 404
  • CLI // Fixed oxen config global setting so that we don’t require a local .oxen repository
  • UI // Text area in Row Details pane should not let content overflow

​
May 7, 2024

​
Feature Updates

  • Improved S3 account validation when linking AWS account
  • Added search for users when adding collaborators
  • Improved large file download in UI
  • Docs// Add HTTP endpoints to public developer docs

​
Fixed Bugs

  • Improved error handling on reading invalid parquet, jsonl, csv, arrow files
  • Fixed search by organization

​
April 26, 2024

​
Feature Updates

  • Improved flow to improve claiming an unclaimed organization
  • Oxen blog now respects syntax highlighting when importing code from ghost
  • Collaborated images are all one fixed image size
  • Added careers link to bottom of mobile website
  • Can now load large data-frames on Oxen server

​
Fixed Bugs

  • Fixed binary in data-frame response on Oxen-server
  • Fixed upload paths for Windows CLI
  • Fixed inaccurate GB storage UI
  • Large files are now downloading fully
  • Saved compare name as typing or on unfocus instead of just enter