Inference vs Training

Inference is running a trained model to generate output; training is building the model's weights in the first place. These mini PCs are inference machines — sized to load and run existing models, not to train new ones from scratch, which needs far more compute and high-bandwidth memory than any unified box provides. Even the 128GB boxes and the M3 Ultra are described for running models (a 70B at Q4, or DeepSeek R1 671B on a 512GB Ultra), not pretraining them. Light fine-tuning of small models is feasible, especially via MLX on Apple Silicon, but serious training belongs on datacenter GPUs.

Buy these boxes to run models, not to train them. If your goal is training or heavy fine-tuning, none of this hardware is the right tool — that's a rented-GPU or datacenter job.