Hello,
We are writing to inform you that we are sunsetting the Model experiments feature in Labelbox on July 15, 2026.
What are Model experiments?
The Experiments feature under the Model tab were designed for:
- Organizing and tracking model runs
- Uploading and comparing model predictions against ground truth labels
- Analyzing model performance with automated metrics
- Conducting error analysis
In short, Model experiments were designed for teams looking to iterate on and systematically improve their ML models directly within Labelbox.
What this means for you
Effective July 15th, the Model experiments feature will no longer be available. After this date:
- You will no longer be able to create new experiments or model runs
- Existing experiment data, including model runs, predictions, and metrics, will be accessible for export until July 15th
- The feature will be fully removed from the platform on July 15th
Note: This update does not impact any of the “Foundry” models or capabilities.
Action items for you
We recommend taking the following steps prior to the sunset date:
- Export any experiment data, model predictions, annotations, and metrics you wish to retain
- Review your current workflows that depend on Model experiments and plan accordingly
- Reach out to your Customer Success Manager or our support team if you have questions about alternative approaches
We understand this change may impact your workflows, and we want to make this transition as smooth as possible. If you have questions or need guidance on exporting your data or adapting your processes, please contact us at support@labelbox.com or reach out to your dedicated Customer Success Manager.
Thank you for being a valued Labelbox customer.
Sincerely,
The Labelbox Team
