AI City Remix$11,500 prize pool. Pick a city, let AI tell its story, Oct 22 – 28.Join now
Start building

Your Agent, ready for real work.

Models, APIs and Runtimes behind one key and one balance. Pay per call, see every run.

Browse the StoreStart with Setup

Three ways to put it to work.

Use it in your local AI tools, call it from your app, or build a reusable Agent. One key covers all three.

  • Setup

    Use in your local AI tools

    Pick useful capabilities from SandBase, run Setup once, and use them in Codex, Claude, Cursor, or Kiro.

    set up https://www.sandbase.ai/SKILL.md

  • API

    Call an API

    Copy a working request for a Model, search, scraping, data, or SaaS API — the price is on the card.

    Browse APIs

  • Agent

    Build an Agent

    Define the job, give it tools, test it, then publish or schedule it.

    Build an Agent

Featured models and X APIs.

Selected LLMs, image and video models, plus X search and profile APIs — with prices from the live catalog.

Browse the Store

Call one capability from your product.

Use a capability directly. Reuse it inside an Agent when the work grows.

Models
Add reasoning, vision, audio, image, and embedding models.
Search APIs
Let your Agent search the live web and return sources.
Data extraction
Turn webpages and files into clean data your Agent can use.
SaaS APIs
Connect to business systems, social data, and workflow tools.
Agent templates
Start from working examples and adapt them to your use case.
Skills
Reuse packaged instructions and code inside Agent workflows.
Browse APIs

Turn a question into finished work.

Research, enrich, and deliver a source-backed brief from one reusable workflow.

  1. 01 Business question

    • Research a company
    • Set the scope
    • List the questions

  2. 02 Research Agent

    • Search live sources
    • Add company data
    • Build the brief

  3. 03 Deliverable

    • Customer brief
    • Sources cited
    • Ready to share

Define it, test it, deliver it.

Give the job a shape, watch one run before anything depends on it, then hand it to an AI tool, an API, or a schedule.

Build an Agent

  1. 01

    Define

    Define the job and give the Agent the right capabilities.

  2. 02

    Test a run

    See the result before anything depends on it.

  3. 03

    Deliver

    Deliver it to an AI tool, an API, or a schedule.

Every run
leaves a record.

Each execution becomes a run with its input, output, trace, and status — and Activities shows what it cost.

Input & output
The request that started it. The result it delivered.
Trace & status
Every step along the way, kept with the run.
Usage & cost
Activities shows what each execution cost.

Real runs, real receipts.

Public runs from SandBase Lab. Each one shows the models it called, how long it took and what it cost.

Explore the Lab ↗

Pay for what you run.

  • Start

    $0

    Explore Models, APIs and Agent workflows with starter credits.

    Start free

  • Build

    Metered

    Pay for the underlying Model, API and runtime work your calls use.

    Create an API key

  • Scale

    Custom

    Centralize workspaces, controls and support for production teams.

    Contact sales

Choose your way in.

Setup puts tools inside your AI app. APIs let your product call one tool directly. Build Agent is for reusable, multi-step work.

SandBase is for builders who need working AI capabilities quickly, and FDEs who need to turn customer requirements into reusable agent workflows.

Use Setup when you want tools inside Codex, Claude, Cursor, or Kiro. Use APIs when your product needs one tool, like search or image generation. Use Build Agent when the job has multiple steps and should be reused.

The Store is where you find reusable agent capabilities: models, APIs, ready-made agents, and skills. You can test them, copy them, install them into Setup, or clone them into Build Agent.

An agent workflow is a reusable job with instructions and tools. It can research a customer, enrich leads, monitor competitors, triage GitHub issues, or run another repeatable process.

You can test it, publish it so apps and AI tools can call it, or schedule it to run automatically. Every execution becomes a run with input, output, trace, and status.

Webhooks notify your system when async model, API, or agent work finishes or fails, so nothing has to poll for the result.

Start with one capability.Finish with work you can hand over.

Explore the Store ›Create a workspace ›