How a model learns: training and inference
Training is the expensive one-time event where a model's numbers get tuned. Inference is the cheap repeated use afterwards. The gap in cost is enormous, and it shapes the whole industry.
Blog
Posts on AI engineering, LLM systems, and software development.
Training is the expensive one-time event where a model's numbers get tuned. Inference is the cheap repeated use afterwards. The gap in cost is enormous, and it shapes the whole industry.