Skip to content

Introduction

Pipelines are built for generic usecases, which combine multiple frameworks to solve a particular problem with a Large Language Model (LLM).

Pipelines workflow: - consist of two or more Frameworks. - could have one or more Datasources. - could have preprocessing and post processing steps on prompts and response from LLM. - is developed by NEXTgpt contributors.

Types of Pipelines

Please check the sub sections for different types of Pipelines and how they can be used.

  1. Document Scoring Pipeline