Vertex AI offers capabilities to customize and tune foundation models on your data for improved performance, as a unique offering to your customers. This includes techniques such as supervised fine-tuning (SFT), reinforcement learning from human feedback, and distillation. We will dive deep into SFT, learn how Vertex lets you customize Gemini models with hundreds of examples cost-effectively. We will also discuss distillation, a technique that uses a teacher model to train smaller student models to perform certain tasks better, at a lower cost and with lower latency. You will also get tips from Palo Alto Networks, based on their real-world experience.

Click the blue “Learn more” button above to tap into special offers designed to help you implement what you are learning at Google Cloud Next 25.