Comparing LLAMA 3, Phi 3, and GPT-3.5 Turbo AI Agents for Web Search Performance
Offered By: Data Centric via YouTube
Course Description
Overview
Explore a comprehensive video comparing the performance of open-source AI models Llama 3 70b, GPT-3.5 turbo, and Phi-3 medium against Perplexity AI in web search tasks. Dive into a detailed benchmarking process using a custom web search agent, examining each model's capabilities through a series of test questions. Learn about the setup and implementation of these models, including how to build your own local "perplexity" with Ollama. Gain insights into the results and effectiveness of each AI model in handling web search queries. Access additional resources for AI development, including a hands-on RAG application project and an AI Engineering course. Stay updated on the latest advancements in AI, Data Science, and Large Language Models through provided links and references.
Syllabus
Introduction:
Test Questions and approach:
Testing GPT 3.5-Turbo:
Results GPT 3.5-Turbo:
Testing Llama3-TenyxChat-70B:
Results Llama3-TenyxChat-70B:
Testing Phi 3 Medium Instruct:
Results Phi 3 Medium Instruct:
Taught by
Data Centric
Related Courses
Fine-tuning Phi-3 for LeetCode: Dataset Generation and Unsloth ImplementationAll About AI via YouTube LLM News: GPT-4, Project Astra, Veo, Copilot+ PCs, Gemini 1.5 Flash, and Chameleon
Elvis Saravia via YouTube LLM Tool Use - GPT4o-mini, Groq, and Llama.cpp
Trelis Research via YouTube LoRA Fine-tuning Explained - Choosing Parameters and Optimizations
Trelis Research via YouTube Building an Orca Mini and Phi 3 Chatbot with Python on Raspberry Pi 5
Arm Software Developers via YouTube