Learn facts. Then query them with natural language.. Associate different responses with contexts that you can filter for.

Useful for saving curated natural language content semantic search.


Steamship Packages are cloud-hosted AI libraries you can use from any programming environment.

Packages manage their own data & infrastructure in the cloud. Each instance gets its own private scope.


Download a Jupyter Demo

> npx try-steamship oi

From Python

from steamship import Steamship

# See GitHub README for Config requirements
instance = Steamship.use("oi", config={})

# See GitHub README for available methods
resp = instance.invoke('method', arg1=val1, arg2=val2)
Pulled from the GitHub repository.
# Steamship Package Template 

This repository contains a starter template showing you how to develop and use a package.

## Quick Start

Deploy and use this package in under a minute!

First, make sure you have the Steamship CLI installed:

npm install -g steamship && ship login

Next, provide a handle for this project in `steamship.json`. Package handles must be globally unique. We recommend `YOURNAME-demo`.

  "handle": "yourname-demo",

Then, deploy your package!

ship deploy

Wait about 30 seconds after deployment finishes for the package to become available.

### Invoke your Package from the CLI

Then, create an instance.

ship package:instance:create --default_name="Beautiful"

That keyword argument above is part of the required configuration.
You can see where it's defined in both the [steamship.json](https://github.com/steamship-packages/oi/blob/main/steamship.json) file and in the [src/api.py](https://github.com/steamship-packages/oi/blob/main/src/api.py) file.

The response will let you know what your **Instance Handle** is.

Finally, invoke a method!

ship package:instance:invoke --instance="INSTANCE_HANDLE" --method="greet"    

### Invoke your Package from Python

It's more likely you'll want to call your package from software you're writing. Let's try from Python.

Create a new instance and invoke it with:

from steamship import Steamship

# TODO: Replace with your package and instance handle below
instance = Steamship.use("PACKAGE_HANDLE", "INSTANCE_HANDLE", config={
    "default_name": "Beautiful"


## Extending on your own

Steamship packages run on a cloud stack designed for Language AI.

You can import files, parse and tag them, query over them, and return custom results. 

Full documentation for developers is available at [https://docs.steamship.com/packages/developing](https://docs.steamship.com/packages/developing).

Steamship is building the fastest way to ship language AI.

Stop building and start shipping. Reach out to get onboarded.
Read More