Multi Modal Support in API?

Watched the opening keynote today and I am confused. Is the GPT 4 Vision not available in the playground? What about images as inputs? The only way I see to add an image as an input is under the new assistants area in the playground. If I go to the “Chat” dropdown there is no option to add an image as an input. Am I doing something wrong?

In my playground, there is no GPT 4 vision, too, so no image upload. However, I can use the API. I already tested it.

1 Like