# From Data to Device: ELMOD An Efficient German-First 2.7B Language Model for Mobile Inference

arXiv: https://arxiv.org/abs/2607.24585 · HTML: https://mlsanity.com/paper/2607.24585/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 41 references that are in the map.
- First posted on arXiv 27 Jul 2026.

## Builds on (most cited first, 26 of 26)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [mT5: A massively multilingual pre-trained text-to-text transformer](https://mlsanity.com/paper/2010.11934/) (2020)
- [Scaling Language Models: Methods, Analysis & Insights from Training Gopher](https://mlsanity.com/paper/2112.11446/) (2021)
- [The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale](https://mlsanity.com/paper/2406.17557/) (2024)
- [Deduplicating Training Data Makes Language Models Better](https://mlsanity.com/paper/2107.06499/) (2021)
- [CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Data](https://mlsanity.com/paper/1911.00359/) (2019)
- [Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus](https://mlsanity.com/paper/2104.08758/) (2021)
- [Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling](https://mlsanity.com/paper/2401.16380/) (2024)
- [mGPT: Few-Shot Learners Go Multilingual](https://mlsanity.com/paper/2204.07580/) (2022)
- [Tokenization counts: the impact of tokenization on arithmetic in frontier LLMs](https://mlsanity.com/paper/2402.14903/) (2024)
- [OLMES: A Standard for Language Model Evaluations](https://mlsanity.com/paper/2406.08446/) (2024)
- [NumeroLogic: Number Encoding for Enhanced LLMs' Numerical Reasoning](https://mlsanity.com/paper/2404.00459/) (2024)
- [Towards Multilingual LLM Evaluation for European Languages](https://mlsanity.com/paper/2410.08928/) (2024)
- [How Does Code Pretraining Affect Language Model Task Performance?](https://mlsanity.com/paper/2409.04556/) (2024)
- [Efficient Language Model Training through Cross-Lingual and Progressive Transfer Learning](https://mlsanity.com/paper/2301.09626/) (2023)
- [Beyond Text Compression: Evaluating Tokenizers Across Scales](https://mlsanity.com/paper/2506.03101/) (2025)
- [Rethinking Multilingual Continual Pretraining: Data Mixing for Adapting LLMs Across Languages and Resources](https://mlsanity.com/paper/2504.04152/) (2025)
- [LL"aMmlein: Transparent, Compact and Competitive German-Only Language Models from Scratch](https://mlsanity.com/paper/2411.11171/) (2024)
- [Scaling Behavior for Large Language Models regarding Numeral Systems: An Example using Pythia](https://mlsanity.com/paper/2409.17391/) (2024)
- [Aleph-Alpha-GermanWeb: Improving German-language LLM pre-training with model-based data curation and synthetic data generation](https://mlsanity.com/paper/2505.00022/) (2025)
- [Pre-Training LLMs on a budget: A comparison of three optimizers](https://mlsanity.com/paper/2507.08472/) (2025)

## Abstract (arXiv)

We present ELMOD - Efficient Language Model for On-Device Deployment - a compact (2.7B) German language model designed for efficient inference on resource-constrained hardware. ELMOD was trained on a limited computational budget (55k H100 GPU hours) using exclusively publicly available data. We developed a suite of German-specific data pre-processing, which differ from English-oriented counterparts in their handling of morphological variation, compounding, and orthographic conventions. Furthermore, we introduced a quality filtering and rephrasing step, which increased the instructional quality of the data, improved performance during the annealing phase, and reduced overall compute requirements. Thanks to our architectural model and data choices, including prefiltering, our educational-quality filtering and rephrasal to raise the educational-quality, ELMOD is the strongest performer in its size class (<3B), matching the performance of 7B-parameter models in German.