📝 Overview
Yoom allows you to integrate with OpenAI to generate text and perform image analysis. In this guide, we'll show you how to set up an OpenAI assistant to analyze images by configuring a flow that processes images uploaded through a form.
☝️ Note:
Using ChatGPT or OpenAI's API may incur additional costs. Please check OpenAI's official website for more details.
🕵️♀️ Case Study
This example demonstrates how to use images and files submitted via a form to obtain answers from a pre-created assistant.
🛠️ How to Configure
1. Create a Custom Assistant
Access the assistant creation screen here.
Create your assistant of choice.
☝️ Note:
Depending on your OpenAI plan, you may have access to newer generative models.
2. Create a Form Trigger in Yoom
Create a new Flowbot in Yoom and set up a form trigger. This form will collect data such as company name, contact person, and images or files.
👉 For more details on form creation, refer to the help page.
3. Upload the Images or Files
Under the trigger, click the "+" sign and choose OpenAI from the integrated app operations.
Configure the following settings:
Action: Upload file
File attachment method: Select "Use output" and choose the images or file data uploaded via the form.
4. Create a Thread in OpenAI
Click the "+" sign, and select OpenAI.
Configure the following:
Action: Create thread (Assistant API)
Test: Click “Test” to obtain the thread ID.
5. Create a Message in OpenAI
Click the "+" sign, select OpenAI, and configure the following:
Action: Create a message (Assistant API)
Thread ID: Embed the thread ID from step 4.
Message content: Enter any prompt or instruction text.
File ID: Embed the file ID from the file upload in step 3.
6. Execute the Assistant
Click the "+" sign, select OpenAI, and configure the following:
Action: Execute the assistant (Assistant API)
Thread ID: Embed the thread ID from step 4.
Assistant ID: Choose the assistant you created.
7. Retrieve the Latest Thread Conversation
Click the "+" sign, select OpenAI, and configure the following:
Action: Retrieve the latest thread conversation (Assistant API)
Thread ID: Embed the thread ID from step 4.
Advanced settings: Enter any number up to 60 for the execution interval.
☝️ Note:
The time delay helps avoid errors caused by execution lags.
🙌 Additional Note
If the delay set in step 7 is insufficient, you can use the “Wait” operation to further delay execution.
☝️ Note:
The “Delay” action is available on Business plans or higher.
Configure the Delay
Click the "+" sign, select the “Delay” action, and configure the following settings:
Setting method: Wait for a specified time
Unit of measurement: This can be set in seconds or minutes.
Time duration: Set as desired.
This concludes how to use an assistant created by OpenAI to analyze images. 🎉
📚 Related Guides
How to Register OpenAI in Yoom - View Guide Here
Basic & Optional Form Settings - View Guide Here
Search Keywords
Search Keywords
OpenAI, ChatGPT, Assistant, Assistant API, GPTs, Image Analysis, Images, Files