​
Tensor One SDKs

The Tensor One SDKs provide powerful tools for interacting with the Tensor One API, enabling integration of custom logic, infrastructure management at scale, and automation of deployment workflows.

​
Interacting with Serverless Endpoints

Once deployed, Serverless Functions are exposed as Serverless Endpoints, which can be called via standard HTTP requests.

​
How It Works

  • Endpoints behave like HTTP APIs
  • Provide your Endpoint ID and API Key to authorize requests
  • Enables seamless interaction between external apps and your deployed logic

​
Infrastructure Management with the SDK

The SDK supports full programmatic control of your cloud environment, including Clusters, Templates, and Serverless Endpoints.

​
Managing Clusters

Clusters are isolated environments for running applications, services, and workloads.

​
Key Actions

  • Create a Cluster: Spin up new Clusters with specified hardware/software configurations
  • Configure Resources: Set GPU type, memory, storage, and network options
  • Deploy Applications: Push custom services or code to run inside Clusters
  • Monitor & Scale: Programmatically track usage and adjust resources as needed

​
Managing Templates and Endpoints

Templates define reusable Cluster configurations. Endpoints expose your services to the outside world.

​
Workflow

  1. Create a Template
    Define base OS, dependencies, container disk, ports, and environment variables
  2. Launch Clusters from Templates
    Consistently spin up pre-configured environments for development or production
  3. Create Serverless Endpoints
    Expose applications running in Clusters via secure HTTP APIs

​
Why Use the SDK?

Using the Tensor One SDKs, you can:
  • Automate deployments
  • Scale infrastructure programmatically
  • Expose intelligent services as secure, serverless APIs
Integrate Tensor One into your CI/CD pipeline or custom backend logic with just a few lines of code.