Customizing Generative AI Models with Vertex AI

Explore how Vertex AI simplifies the customization of generative AI models in this 37-minute conference talk from Google Cloud Next 2024. Dive deep into supervised fine-tuning (SFT) and learn how to cost-effectively customize Gemini models using hundreds of examples. Discover the technique of distillation, which employs teacher models to train smaller student models for improved task performance, reduced costs, and lower latency. Gain valuable insights from Palo Alto Networks' real-world experience as speakers Anand Iyer, May Hu, and Lavanya Ramani discuss Vertex AI's unique offerings for enhancing foundation models with your data, including reinforcement learning from human feedback.