AI RESEARCH

Fine-Tuning and Serving Gemma 4 31B on Google Cloud TPU: A Technical Comparison with GPU Baselines

arXiv CS.AI

ArXi:2605.25645v1 Announce Type: cross We present the first end-to-end nstration of fine-tuning and serving Google's Gemma 4 31B model on TPU hardware, providing an empirical comparison of TPU and GPU platforms for large language model adaptation. Using LoRA on a Google TPU v5p-8 for