> ## Documentation Index
> Fetch the complete documentation index at: https://docs.smlcrm.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Welcome to Simulacrum

> Forecast faster with a Python-first SDK that wraps Simulacrum's time-series foundation models.

# Simulacrum Python SDK

<img src="https://mintcdn.com/simulacrum-68d93327/UkqaErAsBSDE3YAZ/logo/simulacrum-blue.png?fit=max&auto=format&n=UkqaErAsBSDE3YAZ&q=85&s=3a51f687ba2977b94a6b6aad4fc61283" alt="Simulacrum logo" width="512" height="113" data-path="logo/simulacrum-blue.png" />

Simulacrum projects your time series into realistic futures so you can reason about demand, risk, and opportunity with confidence. This documentation walks you through the Python SDK, the underlying HTTP API, and the practices we recommend when you embed forecasting into products or internal workflows.

<Callout type="note" title="Try it locally">
  Install the SDK with `pip install simulacrum-sdk` and follow the quickstart to generate your first forecast in under five minutes.
</Callout>

## What you can ship with Simulacrum

* Fast forecasts backed by Simulacrum's production API.
* Guard rails for quota, authentication, and model selection errors.
* Patterns for backtesting, batched inference, and governance in production.

## First look

```python theme={null}
from simulacrum import Simulacrum

client = Simulacrum(api_key="sim-key_id-secret")
forecast = client.forecast(series=[128.5, 131.7, 133.0, 137.9], horizon=4, model="tempo")
print(forecast.tolist())
```

The guides tab contains end-to-end walkthroughs, while the reference tab documents every class, model, and endpoint exposed by the SDK.

<Steps>
  <Step title="Install" description="Add the `simulacrum-sdk` package to your virtual environment." />

  <Step title="Authenticate" description="Issue a client with your API key and optionally override the base URL." />

  <Step title="Forecast" description="Send clean, single-dimensional series to `/{modelName}/v1/forecast` (default model: `tempo`) and use the numpy response downstream." />

  <Step title="Validate" description="Check key status, quotas, and metadata with a single call to `/{modelName}/v1/validate` (e.g., `/tempo/v1/validate`)." />
</Steps>

Need a hand? Reach out via the support link in the navigation bar or open an issue on GitHub.