Quickstart Guide

​

Why Prem?

Prem is a platform that’s packaged with the necessary tools to effortlessly integrate Generative AI into your applications.

​

With Prem you can:

• Simultaneously test and monitor LLMs in the Prem Lab: Compare the performance of multiple models and find the best fit for your use case.
• Train models in the Prem Gym: Leverage Prem’s ability to autonomously fine-tune models. The Gym is your place to beef up your deployment models by continuously fine-tuning your data in order to improve performance.
• Seamlessly create RAG-based assistants: Prem is capable of interacting with your documents. Just upload the documents, and we’ll take care of the rest.
• Launch and use the Prem JavaScript and Python SDKs: Integrate your custom AI into your applications with minimal effort.

​

Quickstart

✌️ This guide shows you how to create a Generative AI application with Prem in minutes.

1

Activate your Prem account.

Click here to sign up

2

Sign in and generate an API key.

3

Create a new project.

We set up a starter project for you already. You can either use the starter project provided or create a new project from scratch. It’s up to you.

Make sure to take note of your Project ID. You’ll need it to use the SDK. You can find your Project ID on the Project page under each project’s title or on the Set Up page.

4

Head over to the Lab and experiment with various LLMs simultaneously.

5

After you've experimented with different models in the Lab, go to the Launchpad and choose the most suitable model for your use case.

6

Now it's time to integrate Prem into your application.

Choose an SDK and install

Prem also exposes a simple REST API for interacting with the platform.

Python

JavaScript

pip install premai

Run your first project.

Python

JavaScript

from premai import Prem

client = Prem(api_key=YOUR_API_KEY)

project_id = PROJECT_ID

system_prompt = "You're a helpful assistant"

messages = [
    {"role": "user", "content": "Who won the world series in 2020?"},
]

response = client.chat.completions.create(
    project_id=project_id,
    system_prompt=system_prompt,
    messages=messages,
)

print(response.choices)

​

Next steps

• Continue to test LLMs in the Lab.
• Check out the API reference.
• Connect your documents to the LLM to enhance its knowledge.
• Fine-tune your LLMs in the Gym.

​

Read the guides