Cloud storage paths and task creation
Updated at October 24th, 2025
Choosing the right cloud path is essential to ensure secure, efficient, and compliant data uploads within Sama. Whether you’re using the Sama-managed S3 bucket or your organization’s Azure, GCP, or AWS storage, this guide will help you configure your setup and create tasks seamlessly.
|
|
Which cloud provider are you using?
If you want to create tasks in the Sama Hub, follow the instructions in the “Create Task in Sama Hub” tab. If you prefer to upload data from Azure, AWS, or GCP into the Sama S3 bucket, follow the instructions from the “Upload Data to the Sama S3 Bucket” tab.
Create Tasks in Sama Hub
Allowlist your cloud provider (GCP, Azure, AWS). Follow the guide here.
Get your API and project ID
Before getting started, make sure you have:
- An API access key: learn how to get it here.
- Your project ID: contact your Sama representative to request it.
Query Your Task Schema
Once access to your cloud provider (Azure, GCP, or AWS) is configured, query the schema for your project. Learn more here.
Note
If the data contains pre-annotations, make sure you map the “output schema.”
Create Tasks (in Batches)
After retrieving the schema, create your tasks using the batch creation endpoint, following the JSON format described in the guide here.
Check Status of Batch Creation
Once you’ve created your tasks, you can check the status of the batch creation. Read more about this here.
If you receive a response code other than 200, refer to the API documentation for more information.