Quickstart

The Forefront API provides a simple interface for developers to use open-source models in their applications.

This quickstart tutorial will help get your local development environment setup, fine-tune your first model, and start using it in your application.

If you are an experienced developer or want to just dive into using the Forefront API, check out the API reference.

Through this tutorial, you will learn:

How to install the latest SDKs
Some of the basic concepts of the Forefront API
How to fine-tune your first model
How to inference your model

If you run into any challenges or have questions getting started, please contact our team.

1. Create an account

First, create a Forefront account or sign in. Next, go to the API key page and click "Generate API key". Make sure to save this somewhere safe and do not share it with anyone.

2. Install SDK

Forefront provides a custom Node.js and Python library which makes working with the Forefront API simple and efficient. Install the SDK in your preferred language:

pip install forefront

npm i forefront

3. Fine-tune your first model

For the purpose of this tutorial, we'll file-tune Mistral-7B on 1,000 samples from the Open-Hermes-2.5 dataset.

Start by downloading the dataset:

Before continuing, you can alternatively fine-tune via the UI in the Forefront platform.

4. Inference your fine-tuned model

Next steps

Now that you have fine-tuned your first model and ran it, it's time to explore what else is possible:

For more detailed information on our models and the API, see our GPT guide.
Visit the OpenAI Cookbook for in-depth example API use-cases, as well as code snippets for common tasks.
Wondering what OpenAI's models are capable of? Check out our library of example prompts.
Want to try the API without writing any code? Start experimenting in the Playground.
Keep our usage policies in mind as you start building.

Last updated 1 year ago

3. Fine-tune your first model

For the purpose of this tutorial, we'll file-tune Mistral-7B on 1,000 samples from the Open-Hermes-2.5 dataset.

Start by downloading the dataset:

1MB

open-hermes-1000.jsonl

1,000 samples from the Open-Hermes 2.5 dataset by Nous Research

Before continuing, you can alternatively fine-tune via the UI in the Forefront platform.

Next steps

Now that you have fine-tuned your first model and ran it, it's time to explore what else is possible:

For more detailed information on our models and the API, see our GPT guide.

Visit the OpenAI Cookbook for in-depth example API use-cases, as well as code snippets for common tasks.

Wondering what OpenAI's models are capable of? Check out our library of example prompts.

Want to try the API without writing any code? Start experimenting in the Playground.

Keep our usage policies in mind as you start building.