search

head-side-gearLMRP

hashtag
Overview

This document provides comprehensive documentation for LMRP (Language Model Request Processor) components used in the GLChat Backend Standard RAG pipeline. LMRPs are orchestrator modules that wrap prompt builders, LM invokers, and optionally output parsers to perform end-to-end LM invocations in a single process.

LMRP (Language Model Request Processor) is an orchestrator module that combines three key components:

  1. Prompt Builder - Constructs prompts from templates and input data

  2. LM Invoker - Handles the actual Language Model API calls

  3. Output Parser (optional) - Parses and validates LM responses

PreviousDebugging a Hallucinated ResponseNextStandard RAG LMRP

Last updated