Prompts

At the heart of Lleverage's capabilities lies the power of Prompts—succinct instructions or questions directed towards the AI models we support.

The Prompt Node lets you add a prompt into you workflow. Prompts can have dynamic variables that you can feed output from previous nodes or data from you API workflow trigger.

To move from simple prompts to mini features that actually provide value for you and your applications, we offer a variety of constructs that will allow you to finetune and manage prompts.

Prompt Studio

When adding a Prompt Node to your workflow, you can hop into Prompt Studio for prompt construction. You can also launch Prompt Studio stand-alone, where you can pre-build prompts to use later in a workflow. Within Prompt Studio you can compare responses from different models, different versions of the prompt, save prompts and version prompts.

Prompt Templates

To jumpstart building your prompt, Lleverage offers Prompt Templates. These are pre-crafted prompts designed for popular use cases with well decent prompt engineering practices applied. Created by both the Lleverage team and our community, these templates serve as a time-saving resource, ensuring you don't have to start from scratch.

Model Configs

Model configs are a configuration of a particular model. It allows you to select your model provider and tweak settings to guide the generative model's behaviour. You can have multiple model configs per model, choose which one is the default and in Workflows choose the config you want to use for that workflow.