We present Averroes-Q, a 32-billion parameter bilingual Arabic-English large language model fine-tuned entirely on a single Apple M2 Ultra workstation with 192 GB of unified memory. Built upon the Qwen2.5-32B-Instruct architecture, Averroes-Q is the first publicly released Arabic-capable LLM trained exclusively on consumer-grade Apple Silicon hardware. Using parameter-efficient fine-tuning via Low-Rank Adaptation (LoRA) and Apple's MLX framework, we demonstrate that a complete bilingual instruct
1@techreport{hayulalab2026averroesqtraining,
2 title={Averroes-Q: Training a 32-Billion Parameter Bilingual Arabic-English LLM on a Single Apple M2 Ultra Workstation — Hayula Research},
3 author={Hayula AI Lab},
4 year={2026},
5 url={https://huggingface.co/hayulalab/averroes-q-training-paper}
6}
