# A Systematic Characterization of LLM Inference on GPUs

arXiv: https://arxiv.org/abs/2512.01644 · HTML: https://mlsanity.com/paper/2512.01644/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Dec 2025.

## Built on by (most cited first, 8 of 8)

- [TaxBreak: Unmasking the Hidden Costs of LLM Inference Through Overhead Decomposition](https://mlsanity.com/paper/2603.12465/) (2026)
- [Kernel Forge: An Agent Harness for LLM-based Generation and Optimization of CUDA Kernels](https://mlsanity.com/paper/2607.24762/) (2026)
- [Selecting Language Models for Social Science: Start Small, Start Open, and Validate](https://mlsanity.com/paper/2601.10926/) (2026)
- [Just Pass Twice: Efficient Token Classification with LLMs for Zero-Shot NER](https://mlsanity.com/paper/2604.05158/) (2026)
- [LLMSpace: Carbon Footprint Modeling for Large Language Model Inference on LEO Satellites](https://mlsanity.com/paper/2605.05615/) (2026)
- [EnergyLens: Interpretable Closed-Form Energy Models for Multimodal LLM Inference Serving](https://mlsanity.com/paper/2605.10556/) (2026)
- [Threshold-Based Exclusive Batching for LLM Inference](https://mlsanity.com/paper/2606.00516/) (2026)
- [From Caveman to Expert Analyst: Energy Consumption of Variable LLM Tasks](https://mlsanity.com/paper/2608.12350/) (2026)

## Abstract (arXiv)

This work presents a systematic characterization of Large Language Model (LLM) inference to address fragmented understanding. Through comprehensive experiments, we establish a four-dimensional analytical framework: (1) Two-Phase Heterogeneity Observation; (2) Microarchitectural Root Cause Analysis; (3) System Scaling Principles; and (4) Emerging Paradigm Boundaries. Our investigation progresses systematically from observation to foresight: identifying performance phenomena, revealing hardware causes, validating system behavior, and exploring new paradigms. This study not only consolidates a reliable empirical foundation for existing research but also provides new discoveries and practical optimization guidance for LLM inference.