Learn to build a custom AI agent using the Llama 3 70b model deployed on Runpod with vLLM in this 24-minute tutorial. Discover how to create a flexible and scalable AI project compatible with any Hugging Face LLM. Follow along as the video covers the inference server schema, memory requirement determination, server deployment on Runpod, integration with the agent, and a live demonstration of the custom agent. Gain practical insights into AI engineering, large language models, and custom websearch agent development through hands-on explanations and provided resources.
Building a Custom AI Agent with Llama 3 70B and Runpod - Compatible with Hugging Face LLMs
Data Centric via YouTube
Overview
Syllabus
Introduction:
Inference Server Schema:
Determine memory requirements:
Deploying server on Runpod:
Using the inference server with the agent:
Demoing the custom agent:
Taught by
Data Centric
