llmClassificR

The README opens with

A user-friendly toolkit for text classification with Large Language Models in R, from bag-of-words to Jev. It provides robust batch zero-shot classification with Groq LLMs (direct, few-shot, chain-of-thought, JSON, strict JSON Schema, multi-model ensemble, self-consistency, scoring and LLM-as-judge), Jev "System One" classification (TypeSafe AI) directly from R, embeddings (Voyage AI, Jina AI and local HuggingFace models), a do-it-yourself Jev-like classifier from label descriptions, calibration tools, bag-of-words baselines, learning curves and linear probes, retrieval-augmented generation (RAG), synthetic data generation and quality checks, and integration with Exploratory Graph Analysis (EGA) for unsupervised text classification.