Notifications
Clear all
How To
9
Posts
3
Users
0
Reactions
6
Views
Topic starter
I'm utilizing callin.io to process multiple files with Google Gemini AI, but I'm encountering an aggregation issue. The current workflow is structured as follows:
1. Webhook: This module receives multiple files.
2. Iterator: This splits the file array, allowing each file to be processed independently.
3. HTTP & Gemini AI: Each individual file is downloaded and then analyzed.
4. Array & Text Aggregators: These are intended to combine all the outputs from the Gemini AI module.
5. Final Gemini AI Call: This module is supposed to process a single, combined response.
The core problem is that because each file is processed separately, the aggregators are not collecting all the responses before the workflow proceeds. Consequently, the final Gemini AI module receives individual outputs rather than a single, consolidated summary.
I need assistance with the following:
- How to configure the Array Aggregator to wait until all responses have been received.
- Ensuring the Text Aggregator executes only after all responses have been gathered.
- Passing a single, unified text output to the final Gemini AI module.
Any advice on how to adjust this workflow would be greatly appreciated!
Posted : 09/02/2025 10:22 pm
Please set ‘Sequential processing’ to True in the scenario settings.
Posted : 10/02/2025 3:51 am
Topic starter
Hey, thanks for the input, but this didn't change anything. The goal is to have all the initial data run separately and somehow combine them at the end.
Posted : 10/02/2025 10:07 pm
According to your use case, where does the automation break? Is it okay until the HTTP module is?
Posted : 11/02/2025 1:30 am
Topic starter
Everything is functioning correctly; please disregard the error shown in the image. The core issue lies with the final Gemini module, which I intend to use for consolidating all runs into a summary of their respective processes. I'm encountering a problem where the iterator splits the webhook into three distinct runs (which is necessary), and I had assumed the aggregator at the end would combine these runs into a single text output. However, it processes each run individually to completion, without providing an option to aggregate them in the final Gemini module.
Posted : 11/02/2025 12:10 pm
Hello,
From the webhook, if all the files are not combined into a single array, use the Array aggregator after the webhook. Then, feed that aggregated data into an Iterator.
The final Array aggregator should select a suitable Source Module. Combining data from the source module into the aggregation is crucial.
Check this topic and its replies.
P.S.: Always search first. Check the callin.io Academy. If this was helpful, Mark as Best Answer
and give
If you need expert assistance or have questions, feel free to contact us or comment below!
Posted : 11/02/2025 1:33 pm
Hello,
Your Webhook data is an array within the collection. You don't need to use an aggregator after Webhook. Directly add/select the property_image_uploads array as the source for the Iterator.
Also, try selecting the Iterator Module as the source module for the last Aggregator.
I hope this helps.
Posted : 12/02/2025 2:18 am
Topic starter
Thanks, linking the array aggregator to the iterator accomplishes the task.
Posted : 12/02/2025 1:17 pm