# The Compression Paradox in LLM Inference: Provider-Dependent Energy Effects of Prompt Compression

arXiv: https://arxiv.org/abs/2603.23528 · HTML: https://mlsanity.com/paper/2603.23528/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Prompt and context compression, one of 12 directions in Efficient LLM inference and serving.
- Builds on 23 of its 35 references that are in the map.
- First posted on arXiv 6 Mar 2026.

## Placed in

- Efficient LLM inference and serving / [Prompt and context compression](https://mlsanity.com/inference/prompt-and-context-compression/)

## Builds on (most cited first, 23 of 23)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Energy and Policy Considerations for Deep Learning in NLP](https://mlsanity.com/paper/1906.02243/) (2019)
- [Green AI](https://mlsanity.com/paper/1907.10597/) (2019)
- [Carbon Emissions and Large Neural Network Training](https://mlsanity.com/paper/2104.10350/) (2021)
- [Quantifying the Carbon Emissions of Machine Learning](https://mlsanity.com/paper/1910.09700/) (2019)
- [LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression](https://mlsanity.com/paper/2310.06839/) (2023)
- [LLMLingua-2: Data Distillation for Efficient and Faithful Task-Agnostic Prompt Compression](https://mlsanity.com/paper/2403.12968/) (2024)
- [LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models](https://mlsanity.com/paper/2310.05736/) (2023)
- [Estimating the Carbon Footprint of BLOOM, a 176B Parameter Language Model](https://mlsanity.com/paper/2211.02001/) (2022)
- [Towards the Systematic Reporting of the Energy and Carbon Footprints of Machine Learning](https://mlsanity.com/paper/2002.05651/) (2020)
- [Sustainable AI: Environmental Implications, Challenges and Opportunities](https://mlsanity.com/paper/2111.00364/) (2021)
- [Power Hungry Processing: Watts Driving the Cost of AI Deployment?](https://mlsanity.com/paper/2311.16863/) (2023)
- [The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrink](https://mlsanity.com/paper/2204.05149/) (2022)
- [LLMCarbon: Modeling the end-to-end Carbon Footprint of Large Language Models](https://mlsanity.com/paper/2309.14393/) (2023)
- [Energy Usage Reports: Environmental awareness as part of algorithmic accountability](https://mlsanity.com/paper/1911.08354/) (2019)
- [Carbon-Aware Computing for Datacenters](https://mlsanity.com/paper/2106.11750/) (2021)
- [TokenPowerBench: Benchmarking the Power Consumption of LLM Inference](https://mlsanity.com/paper/2512.03024/) (2025)
- [The Perplexity Paradox: Why Code Compresses Better Than Math in LLM Prompts](https://mlsanity.com/paper/2602.15843/) (2026)

## Abstract (arXiv)

The rapid proliferation of Large Language Models has created an environmental paradox: the very technology that could help solve climate challenges is itself becoming a significant contributor to global carbon emissions. We test whether prompt compression improves inference energy efficiency in 28,421 successful API trials (28,428 planned) across three providers (OpenAI GPT-4o-mini, Anthropic Claude-3.5-Sonnet, and DeepSeek-Chat), five benchmarks (HumanEval, MBPP, GSM8K, MATH, MMLU), and four compression ratios (r in {1.0, 0.7, 0.5, 0.3}). Energy is estimated with a token-based proxy calibrated against local direct measurements, and quality is tracked with benchmark pass rates. Compression produced substantial quality loss (overall pass rate 26.0% at baseline vs. 1.5% at r=0.7) and strongly provider-dependent energy behavior. DeepSeek exhibited output expansion under compression (21 to 798 tokens at r=0.3), corresponding to energy increases up to +2,140%, while GPT-4o-mini showed mixed effects including a reduction at r=0.5. These results indicate that input-token reduction alone is not a reliable energy optimization strategy in production inference. For the evaluated settings, model selection and output-length control provided more consistent energy-quality tradeoffs than prompt compression.