How Small Can You Go? Compact Language Models for On-Device Critical Error Detection in Machine Translation

Large Language Models (LLMs) excel at evaluating machine translation (MT), but their scale and cost hinder deployment on edge devices and in privacy-sensitive workflows. We ask: how small can you get while still detecting meaning-altering translation errors? Focusing on English → German Critical Error Detection (CED), we benchmark sub-2 B models (LFM2-350M, Qwen-3 0.6B/1.7B, Llama-3.2-1B-Instruct, Gemma-3-1B) across WMT21, WMT22, and SynCED-EnDe 2025. Our framework standardizes prompts, applies lightweight logit-bias calibration and majority voting, and reports both semantic quality (MCC, F1-ERR/F1-NOT) and compute metrics (VRAM, latency, throughput). Results reveal a clear sweet spot around one billion parameters: Gemma-3-1B provides the best quality-efficiency trade-off, reaching MCC=0.77 with F 1 -ERR =0.98 on SynCEDEnDe 2025 after merged-weights fine-tuning, while maintaining 400 ms single-sample latency on a MacBook Pro M4 Pro (24 GB). At larger scale, Qwen-3-1.7B attains the highest absolute MCC (+0.11 over Gemma) but with higher compute cost. In contrast, ultra-small models (<0.6 B) remain usable with few-shot calibration yet under-detect entity and number errors. Overall, compact, instruction-tuned LLMs-augmented with lightweight calibration and small-sample supervision, can deliver trustworthy, on-device CED for MT, enabling private, low-cost error screening in realworld translation pipelines. All datasets, prompts, and scripts are publicly available at our GitHub repository.11The code and instructions are available at: https://github.com/AppliedMachineLearning-Lab/How-small-can-you-get