{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/inference-economics-of-language-models","title":"Inference economics of language models","arxiv_id":"2506.04645","date":"2025-06-05","proceeding":null,"authors":["Ege Erdil"],"abstract":"We develop a theoretical model that addresses the economic trade-off between cost per token versus serial token generation speed when deploying LLMs for inference at scale. Our model takes into account arithmetic, memory bandwidth, network bandwidth and latency constraints; and optimizes over different parallelism setups and batch sizes to find the ones that optimize serial inference speed at a given cost per token. We use the model to compute Pareto frontiers of serial speed versus cost per token for popular language models.","url_abs":"https://arxiv.org/abs/2506.04645v1","url_pdf":"https://arxiv.org/pdf/2506.04645v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"inference-economics-of-language-models","repo_url":"https://github.com/ege-erdil/inference-economics","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}