In this article
Albato AI is powered by the built-in GPT-5 model developed by OpenAI. The model is integrated by default, so there’s no need to select it manually.
Albato AI helps process data inside your automations and handles a wide range of tasks, for example, translating and generating texts, code, and product descriptions, as well as parsing emails.
How Albato AI works
You don’t need an OpenAI account to get started. Simply add the Albato AI tool to your automation and configure it. The tool is available on Albato paid plans starting from Basic.
Albato AI charges 1 transaction for every 2,000 tokens generated in the model’s response. You can view the exact number of tokens used in the Automation History when you open the result of the Albato AI step.
The token calculation always rounds up. This means:
If a response contains 1,800 tokens, Albato AI deducts 1 transaction.
If a response contains 2,100 tokens, Albato AI deducts 2 transactions.
A token is a fragment of a word used for text processing. In English, 1 token ≈ 4 characters or about 0.75 of a word. You can learn more about tokens in the article What are tokens and how to count them. Additional examples are available in the OpenAI Cookbook, Playground and other sections on the OpenAI website.
Example automation
Let’s have a look at an automation example where you use Slack and Albato AI to automatically translate messages into Spanish.
Imagine you have Spanish-speaking colleagues who need to receive important updates on a regular basis. You write a message in a shared Slack channel in English, and you want it to automatically appear in another channel in Spanish, without manual copying and external translation tools.
-
Log in to your Albato account and go to the My automations section.
-
Select the trigger you want to use. In this example, the automation will run every time a new message appears in the selected Slack channel.
-
Next, connect your Slack account and grant Albato access to the required channel. You can find the connection instructions by this link.
-
To receive translated messages from a specific channel, set up a filter on the trigger so it processes only messages from that channel.
- To select the Albato AI tool, click the + icon and then choose Action.
- Then click Continue.
- In the window that opens, configure AI parameters for your task and fill in the required fields.
The Role field defines from whose perspective the message is sent to the model and sets the context for the AI (system, user, or assistant). In this case, you should enter user.
In the Message Text field, specify what exactly the AI should perform or process.
This is the field where you describe what you want the AI model to do. It's a free-form text that describes the task. A good description should be clear and specific, providing all the necessary details so that the AI model can complete the task effectively.
Here is the prompt used in our example:
Below is a text from Slack. Translate it into Spanish, preserving its meaning, tone, and style.
- Optional fields should remain empty. Filling them in will cause an error.
- After that, add another Slack action to the automation so you can send the translated message to the channel.
- Configure the fields and choose the channel where you want to receive the translated message.
- Here is what the automation will look like.
All that’s left is to run the automation.
Please note: the tool is available only on Albato paid plans. It does not work on the Trial plan. If you have any issues, contact support via our website’s live chat.
