Overview
Explore the latest OpenAI models, including text-embeddings-small, text-embeddings-large, and GPT-4 Turbo Preview (0125), in this comprehensive 16-minute video tutorial. Learn how to perform multiple embedding calls concurrently using asyncio and make simultaneous GPT calls with threadpool executor. Compare the performance of GPT-3.5-turbo-1106 and GPT-4-0125 models in both JSON and regular modes. Discover techniques for embedding entire books using different models and apply cosine similarity for text comparison. Gain practical insights into working with these cutting-edge AI models through hands-on demonstrations and real-world examples.
Syllabus
INTRO
gpt call class
initializing the gpt class
first test gpt 1106 vs 0125
json mode testing 1106 vs 0125 with 10 calls simultaneously
regular mode testing 1106 vs 0125 with 10 calls simultaneously
testing the mdoels with multiple questions
embedding a book with embeddings large, small and ada-002
cosine similarity
cosine similarity with chat
Taught by
echohive