Explore the world of OpenAI Embeddings in this informative 16-minute video. Delve into the recent launch of OpenAI's embeddings endpoint, examining its potential for text similarity, text search, and code search applications. Analyze the controversy surrounding the quality and cost-effectiveness of these embeddings compared to open-source alternatives. Learn about the available embedding options, evaluate OpenAI's performance claims, and understand the criticisms raised by experts. Gain insights into result discrepancies, author responses, and real-world implications. Conclude with a discussion on OpenAI's pricing strategy and its impact on the AI community.
Overview
Syllabus
- Intro
- Sponsor: Weights & Biases
- What embeddings are available?
- OpenAI shows promising results
- How good are the results really?
- Criticism: Open models might be cheaper and smaller
- Discrepancies in the results
- The author's response
- Putting things into perspective
- What about real world data?
- OpenAI's pricing strategy: Why so expensive?
ERRATA: At I say "better", it should be "worse"
Taught by
Yannic Kilcher
Related Courses
-
Introduction to Embeddings with the OpenAI API
-
Learn Embeddings and Vector Databases
-
Vector Embeddings Tutorial – Create an AI Assistant with GPT-4 & Natural Language Processing
4.5 -
Building a Custom OpenAI Chatbot
-
Mastering OpenAI Python APIs: Unleash ChatGPT and GPT4
-
Cohere vs. OpenAI Embeddings - Multilingual Search
