OpenAI Updates ChatGPT 4! New GPT-4 Turbo with Vision API Generates Responses Based on Images

OpenAI has recently launched GPT-4 Turbo with Vision, a significant enhancement to its AI capabilities. This advanced version of GPT-4 Turbo now integrates vision functionalities, allowing it to process and interpret both text and image inputs. With the introduction of JSON mode and function calling for vision requests, this update makes the model more versatile and user-friendly.
SUBSCRIBE for more! 👉 bit.ly/3zlUmiS 👈
automate everything. 👇
linktr.ee/webcafe
GPT-4 Turbo and GPT-4
platform.openai.com/docs/models/gpt-4-turbo-and-gpt-4
AI Newsletter [FREE] ☕
aitraining.webcafeai.com/joinus
-------------------------------------------------
➤ Follow @webcafeai
• 🏄 2nd Channel: www.youtube.com/@corbinwander
• 🐦 Twitter: webcafeai
• ⛰️ TikTok: www.tiktok.com/@webcafeai
• 🥾 Instagram: webcafeai
• 🎧 Bräunlich: soundcloud.com/braunlich
-------------------------------------------------
Key Takeaways:
✩ Vision Integration: GPT-4 Turbo now includes vision capabilities, enabling it to understand and generate responses based on image inputs, alongside text.
✩ Enhanced Accessibility: The update introduces JSON mode and function calling for vision requests, simplifying the integration and utilization of the model in various applications.
0:30 - Introduction to OpenAI's latest update
1:48 - Examples of applications using the new endpoint
3:20 - Cost comparison with industry competitor Anthropic
▼ Extra Links of Interest:
🤖 AI Courses
aitraining.webcafeai.com/store
💰 Earn 50% Commission on Webcafe Software: webcafesoftware.bixgrow.com/register
My name is Corbin, an AI developer entrepreneur behind the vision of Webcafe AI. Together we will build digital ecosystems. ☕

[@webcafeai]

Navigate to key moments👇 made via tubestamp.com 0:30 - Introduction to OpenAI's latest update 1:48 - Examples of applications using the new endpoint 3:20 - Cost comparison with industry competitor Anthropic Recap by TubeStamp ✏️

[@youadsinfo5422]

Hi, can the assistant output for example real estate information and real estate photos in response?

[@JJSeattle]

When using Ai for developing, helping, etc. I recommend whatever it might be (coding, app, web, writing, product development, sciences, etc.) know at least the very basics of that interest, trust, you'll zoom zoom zoom

[@webcafeai]

Agreed, fundamentals are very important!

[@JJSeattle]

@webcafeai Yeah, even knowing their basic definitions, for sure.

[@tutorialsEU]

Thanks for the update

[@nicolassuarez2933]

Images not working for Node JS... Maybe a bug?

[@vVthinkhardVv]

So is this available via the Assistants? As in can I now upload an image to an Assistant and it will be able to view it, if I choose the GPT4 Turbo model for my Assistant?

[@nickmason5654]

I have been trying to get it to work via assistants and have not had any success. There's two designations - gpt-4-turbo-2024-04-09 and gpt-4-turbo-preview. The second is an alias for the first. When I give it an image in the playground, it describes the properties of the file, but not the actual image that the file is. Not sure what's going on, but right now it doesn't seem like it works.

[@vVthinkhardVv]

@@nickmason5654 Appreciate the update. @webcafeai would be great to get more info on how to use Vision alongside Assistants?

[@webcafeai]

Hey! Using vision directly in the Assistant API is not currently available. However, there is a great workaround. In your logic flow, I would set up a call to the endpoint of gpt-4-turbo-2024-04-09, have it output the relevant vision data you care about, then provide this data to the relevant Assistant API endpoint, and have that endpoint provide your end value. Hope this helps!

[@daviddad7388]

I don't get it. I have had access to vision since last November or October. Am I missing something?

[@TheChromePoet]

I still don't know what the hell is going on. I'm trying to figure out if the chatgpt app got this update.

[@nickmason5654]

as i read it, it's supposed to be the release version of that October/November one. But I can't get the image part of it working so I'm still using gpt-4-vision-preview

[@BradsPitt-bu5fq]

I just spent 2 weeks on this crap and someone should make a video on the Openai migations.

[@DaveThomson]

Hotdog / not hotdog

[@webcafeai]

Great Silicon Valley reference 😂

[@absolutwp6496]

😂😂 3:18

[@youngleon9650]

still not as good as claude 3 especially on the illusion issues

[@negvey]

I hate Claude, I was so hyped to try it when I heard it beat gpt 4... first of all it's not anymore better and second it's response is harder to naturally read and digest, it's like I can't follow word to word really easily, gpt is just a better author, that is my opinion tho, I would t argue it to the teeth