# A Picture is Worth a Thousand Tokens: How Vision Language Models Cut AI Energy Costs While Improving Accuracy

arXiv: https://arxiv.org/abs/2608.07427 · HTML: https://mlsanity.com/paper/2608.07427/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 15 of its 23 references that are in the map.
- First posted on arXiv 7 Aug 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Visual token pruning and efficiency](https://mlsanity.com/mllm/visual-token-pruning-and-efficiency/)

## Builds on (most cited first, 15 of 15)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Pixtral 12B](https://mlsanity.com/paper/2410.07073/) (2024)
- [Tokenization counts: the impact of tokenization on arithmetic in frontier LLMs](https://mlsanity.com/paper/2402.14903/) (2024)
- [Zeus: Understanding and Optimizing GPU Energy Consumption of DNN Training](https://mlsanity.com/paper/2208.06102/) (2022)
- [How Hungry is AI? Benchmarking Energy, Water, and Carbon Footprint of LLM Inference](https://mlsanity.com/paper/2505.09598/) (2025)
- [A Picture is Worth A Thousand Numbers: Enabling LLMs Reason about Time Series via Visualization](https://mlsanity.com/paper/2411.06018/) (2024)
- [The ML.ENERGY Benchmark: Toward Automated Inference Energy Measurement and Optimization](https://mlsanity.com/paper/2505.06371/) (2025)
- [TokenPowerBench: Benchmarking the Power Consumption of LLM Inference](https://mlsanity.com/paper/2512.03024/) (2025)
- [From Prompts to Power: Measuring the Energy Footprint of LLM Inference](https://mlsanity.com/paper/2511.05597/) (2025)
- [Harnessing Vision-Language Models for Time Series Anomaly Detection](https://mlsanity.com/paper/2506.06836/) (2025)
- [Energy-Driven Adaptive Visual Token Pruning for Efficient Vision-Language Models](https://mlsanity.com/paper/2603.05950/) (2026)
- [How Different Tokenization Algorithms Impact LLMs and Transformer Models for Binary Code Analysis](https://mlsanity.com/paper/2511.03825/) (2025)
- [Causal Intervention Sequence Analysis for Fault Tracking in Radio Access Networks](https://mlsanity.com/paper/2511.17505/) (2025)
- [Edge-Deployable LLM Fine-Tuning on a Single GPU for Telecom Network Troubleshooting](https://mlsanity.com/paper/2607.02523/) (2026)

## Abstract (arXiv)

LLM inference accounts for over 90% of AI operational energy, scaling directly with input token count---a critical inefficiency for telecom network analytics and numerical time-series data analysis (NTSDA), where raw multivariate KPI windows from 4G/5G cell sites expand into thousands of floating-point tokens. Vision-Language Models (VLMs) eliminate this mismatch by encoding time-series as 2D plots, achieving 3.6-10.4x input token reduction across Llama-3.2-90B, Qwen2.5-VL-72B, and Pixtral-12B architectures. This translates to 1.8-2.5x measured inference energy reduction, saving approximately 7.2 MJ/day at telecom edge deployments and CloudRAN that monitor 200 cells per 15-minute interval. Critically, efficiency gains do not sacrifice accuracy: a fine-tuned Llama-3.2-90B-Vision VLM achieves 220.7% higher precision than its text-only counterpart and outperforms LSTM and ARIMA baselines by over 144% on telecom anomaly detection. On public benchmarks, Pixtral-12B achieves a 20.6x improvement in J/F1 score at mean F1 = 0.82. At 24 KPIs, text representations exceed the 128K context window of most production LLMs, rendering text-only processing infeasible without truncation, while visual representations remain within standard limits. These results establish VLMs as an energy-efficient and accuracy-superior modality for numerical time-series workloads, providing empirical grounding for AI inference systems that treat energy consumption as a first-class engineering constraint.