# Analysis of LLM Performance on AWS Bedrock: Receipt-item Categorisation Case Study

arXiv: https://arxiv.org/abs/2604.01615 · HTML: https://mlsanity.com/paper/2604.01615/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 32 references that are in the map.
- First posted on arXiv 2 Apr 2026.

## Builds on (most cited first, 9 of 9)

- [A Survey of Large Language Models](https://mlsanity.com/paper/2303.18223/) (2023)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [Best Practices for Text Annotation with Large Language Models](https://mlsanity.com/paper/2402.05129/) (2024)
- [TnT-LLM: Text Mining at Scale with Large Language Models](https://mlsanity.com/paper/2403.12173/) (2024)
- [Knowledge Distillation in Automated Annotation: Supervised Text Classification with LLM-Generated Training Labels](https://mlsanity.com/paper/2406.17633/) (2024)
- [Enhancing Text Classification through LLM-Driven Active Learning and Human Annotation](https://mlsanity.com/paper/2406.12114/) (2024)
- [Evaluating Students' Open-ended Written Responses with LLMs: Using the RAG Framework for GPT-3.5, GPT-4, Claude-3, and Mistral-Large](https://mlsanity.com/paper/2405.05444/) (2024)
- [Easy Mobile Meter Reading for Non-Smart Meters: Comparison of AWS Rekognition and Google Cloud Vision Approaches](https://mlsanity.com/paper/1910.12617/) (2019)
- [Visualisation for the CIS benchmark scanning results](https://mlsanity.com/paper/2512.11316/) (2025)

## Abstract (arXiv)

This paper presents a systematic, cost-aware evaluation of large language models (LLMs) for receipt-item categorisation within a production-oriented classification framework. We compare four instruction-tuned models available through AWS Bedrock: Claude 3.7 Sonnet, Claude 4 Sonnet, Mixtral 8x7B Instruct, and Mistral 7B Instruct. The aim of the study was (1) to assess performance across accuracy, response stability, and token-level cost, and (2) to investigate what prompting methods, zero-shot or few-shot, are especially appropriate both in terms of accuracy and in terms of incurred costs. Results of our experiments demonstrated that Claude 3.7 Sonnet achieves the most favourable balance between classification accuracy and cost efficiency.