Documentation

An outcome-focused guide to what you can do with Howlet Studio, and how easily.

Reference manual (technical documentation)

Quick Start Guide

Get your first AI application running in minutes with this step-by-step guide.

Prerequisites: For Cloud/SaaS, you need a Howlet Studio account via portal.howlet.ai. For on-premise, you need Docker 24+, PostgreSQL 16+, and minimum 8 CPU cores with 32 GB RAM.

Step 1: Access Your Workspace

Log in to the Customer Portal at portal.howlet.ai. Navigate to Tenants and click Open Workspace on your assigned environment.

Step 2: Create Your First Flow

In the workspace, click New Flow in the top bar. Select the "Simple Chat" template to start with a pre-configured conversational AI flow.

Step 3: Configure the LLM Node

Click the LLM node in your flow. Select your model provider, enter your API key, and configure temperature and max tokens.

Step 4: Test with Playground

Click the Playground button in the bottom-right. Type a message and verify the AI responds correctly.

Tip: Iterate on your prompt template until you're satisfied with the output quality. Small tweaks can make a big difference.