Vision Models (GPT-4o, GPT-4 Turbo Vision)
This article provides instructions for accessing and using Vision Models on Powerflow while logged in. If you're unsure of how to log in, follow this guide.
The GPT-4o and GPT-4 Turbo Vision are the GPT models that allows users to upload or import images for use in their prompts.
Step 1
After logging in, navigate to your dashboard. Click on "Powerflow." If you haven't installed Powerflow yet, you can do so from the App Store.
For guidance on how to install apps, refer to this guide.
Step 2
Ensure you are in Stream Mode by selecting "Stream" from the options next to the model selector, as depicted in the image above.
Stream Mode is the default mode on Powerflow, and only this mode can use the vision features of the models GPT-4o and GPT-4 Turbo Vision. Although, Vision features are available only on stream mode, the GPT-4o can be used to parallel and sequential without Vision features.
Step 3
From the Dashboard, click on the Powerflow app listed under frequently used apps. In the Powerflow app, click on the model dropdown menu and select the GPT-4o or the GPT-4 Turbo Vision Model as depicted in the images above.
The primary differences between GPT-4o and GPT-4 Turbo Vision lie in performance and cost. GPT-4 Turbo Vision is designed to be faster and more cost-effective than GPT-4o, while maintaining similar capabilities in terms of language understanding and generation.
Step 4
Under the message input bar, you have three icons/options for uploading images as highlighted in the above image:
Related Topics
Back to top