Newest papers

How Developers Configure LLM Inference: A Large-Scale Study of Real-World Practices

Nada Zine, Matias Martinez, Romain Rouvoy, Clément Quinton · 2026 · arXiv

DOI: 10.48550/arXiv.2610.27945 · open-access PDF

Portuguese

Os LLMs estão cada vez mais integrados em sistemas de software, tornando a implantação e a configuração de inferência uma tarefa de engenharia de software recorrente. Os backends de inferência expõem muitas opções que afetam o carregamento do modelo, a geração, a prestação de serviços e o processamento de solicitações. No entanto, pouco se sabe sobre qual dessas opções os desenvolvedores explicitamente configuram na prática. Neste artigo, apresentamos um estudo empírico em larga escala das práticas de configuração de inferência do LLM em software de código aberto usando os Hugging Face Transformers, o vLLM e a API OpenAI. Analisamos 5.398 repositórios do GitHub usando pelo menos um desses backends

Machine translation (Meta M2M100); the English original follows.

Abstract

LLMs are increasingly integrated into software systems, making deployment and inference configuration a recurring software engineering task. Inference backends expose many options that affect model loading, generation, serving, and request processing. Yet little is known about which of these options developers explicitly configure in practice. In this paper, we present a large-scale empirical study of LLM inference configuration practices in open-source software using Hugging Face Transformers, vLLM, and the OpenAI API. We analyze 5,398 GitHub repositories using at least one of these backends

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.27945, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange