Explore the world of open source LLMs and multi-modal foundation models for developing cost-effective and flexible domain-specific generative AI applications in this 21-minute talk. Gain insights into optimizing models for AI agent use cases, including application copilots and domain-specific assistants that leverage tools. Discover best practices for fine-tuning models to enhance retrieval, chat, planning, and function calling capabilities. Learn about the open source foundation model ecosystem and its potential for creating powerful AI agents. Presented by Tristan Zajonc, Cofounder & CEO of Continual, this talk offers valuable knowledge for those interested in advancing their AI development skills using open source technologies.
Overview
Syllabus
Fine-Tuning Open Source Models for Agents
Taught by
Databricks