Does gpt-4-1106-preview include Gpt-4V?

API
, ,
1

Hi, I’ve read in the documentation that GPT-4 users get GPT Vision in the API and they should be referred as ‘GPT-4V’ or ‘gpt-4-vision-preview’

I see the following options:

  1. gpt-4-1106-preview
    1. gpt-4-0613
  3. gpt-4-0314
  4. gpt-4

Are any of those Vision capable modules?

2

gpt-4-vision-preview is the gpt-4 model with vision capabilities enabled.

Here’s an excerpt from docs:

It is important to note the following:

  • GPT-4 with vision is not a model that behaves differently from GPT-4, with the small exception of the system prompt we use for the model

  • GPT-4 with vision is not a different model that does worse at text tasks because it has vision, it is simply GPT-4 with vision added

  • GPT-4 with vision is an augmentative set of capabilities for the model

2 Likes
3

Thanks for replying to this thread, I just can’t seem to work out the Vision capabilities in the Playground so I was wondering maybe I don’t have it at all, since the documentation says here:

GPT-4 with Vision, sometimes referred to as GPT-4V or gpt-4-vision-preview in the API, allows the model to take in images and answer questions about them.

GPT-4 with vision is currently available to all developers who have access to GPT-4 via the gpt-4-vision-preview model and the Chat Completions API which has been updated to support image inputs.

Since my model dropdown doesn’t explicitly say “GPT-4V” or “gpt-4-vision-preview”, do I even have that option?

4

Yes, you absolutely have access to vision if you have gpt-4. As of writing this post, the playground only supports text inputs for chat completion and completion API.

Hence, the best way to try the vision model is to simply use the list models endpoint to check the models available to you and use the boilerplate code in the documentation I shared to access the vision model.