SandBase Docs

Build with Models and APIs directly, or define reusable Agents that run on managed infrastructure.

SandBase provides two complementary ways to build AI applications. Start with the smallest surface that fits the work; you can combine them as the application grows.

Choose how to build

	Models and APIs	Managed Agents
What it is	Direct access to language, image, video, audio, embedding, and third-party APIs	Reusable, versioned Agent configurations executed by SandBase
Best for	Custom application loops and fine-grained request control	Multi-step, asynchronous, scheduled, or repeatable work
You manage	Prompts, orchestration, retries, and application state	Agent definition, enabled capabilities, inputs, and delivery behavior
Start here	Models and APIs	Build Agent

One SandBase API key

The same API key can call Models, APIs, Agents, Endpoints, and scheduled workloads. Keep keys on the server and never expose them in browser code.

Core concepts

Models and APIs

Use Models for language, image, video, audio, and embedding workloads. Use APIs when the application needs external data or actions. The Store shows available capabilities, pricing, input schemas, and working examples.

• Browse the Store
• Supported Models
• Make your first API call
• Open the API Reference

Agents

An Agent is a reusable, versioned configuration containing a Model, instructions, APIs, Skills, and credentials. Define it once, test it, and then choose how it should run.

• Understand Agents
• Define an Agent
• Add APIs, Skills, and credentials

Services and Sessions

A Service exposes a tested Agent to applications. Its API resource is an Endpoint. Invoking it creates or continues a Session whose session_id identifies the durable interaction.

• Services guide
• Service quickstart
• Endpoints API
• Sessions API

Schedules, DeploymentRuns, and Sessions

Schedules execute an Agent later or repeatedly. Their API resource is a Deployment. Every trigger creates a distinct drun_* DeploymentRun and, on successful Session creation, links it to a new sess_* Session.

• Schedules guide
• Deployments API
• Sessions guide

First steps

1. Create an organization and API key.
2. Choose a Model, API, or Agent from the Store.
3. Make a direct request or define and test an Agent.
4. Publish the Agent as an Endpoint when an application needs to invoke it.
5. Add a Schedule when the work should run automatically.
6. Review Sessions, errors, latency, and cost in the Console.

Operational guidance

• Streaming responses
• Error handling and retries
• Rate limiting
• Billing and pricing
• Workspace administration

Next steps

• New to SandBase: Getting started
• Calling an interface: API Reference
• Building reusable work: Build Agent
• Connecting an AI tool: Setup