From the course: GPT-4 Turbo: The New GPT Model and What You Need to Know
Unlock the full course today
Join today to access over 23,700 courses taught by industry experts.
GPT-4 Turbo with vision - GPT Tutorial
From the course: GPT-4 Turbo: The New GPT Model and What You Need to Know
GPT-4 Turbo with vision
- [Instructor] Language models have normally only had text as input. The GPT-4 vision model allows you to take in either text or images or both as input and produce text as output. This allows you to ask questions about the images because the model has an excellent understanding of the images. This is different from something like DALL.E 3, which allows you to provide text as input and generates images. So let's go ahead and try this out. Now, this time around, I'm using ChatGPT+. Now, ChatGPT+ is a subscription plan offered by OpenAI for the ChatGPT service. I'm using this because at the moment, there isn't an easy way to upload a document in the OpenAI playground. So let me go ahead and grab an image. I'm going to be using an image of myself sitting in an office environment as input. And let's ask the question, "What's in this image?", and send it to the model. And you can see the response back from the model is…
Contents
-
-
-
(Locked)
What's new?1m 31s
-
(Locked)
Assistants API3m 8s
-
(Locked)
Code Interpreter4m 50s
-
(Locked)
Code Interpreter tool using the API3m 51s
-
(Locked)
Retrieval tool using the platform4m 9s
-
(Locked)
Retrieval tool using the API1m 35s
-
(Locked)
Function calling using the API7m 41s
-
(Locked)
Function calling on the platform3m 42s
-
(Locked)
Parallel function calling on the platform2m 40s
-
(Locked)
GPTs3m 47s
-
(Locked)
GPT-4 Turbo with vision4m 19s
-
(Locked)
GPT-4 Turbo with 128K context4m 34s
-
(Locked)
-
-