Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2506.04645 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912414873681920 |
|---|---|
| author | Erdil, Ege |
| author_facet | Erdil, Ege |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_04645 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Inference economics of language models Erdil, Ege Machine Learning Distributed, Parallel, and Cluster Computing 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. |
| title | Inference economics of language models |
| topic | Machine Learning Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2506.04645 |