Albato AI is a tool that combines the APIs of various AI platforms, such as OpenAI, ChatGPT, and Stable Diffusion, making their use even more convenient and accessible. With Albato AI, you can automate your processes using AI without the need to register multiple accounts on different platforms.
Currently, the only available platform in Albato AI is OpenAI with the GPT-3.5 version, which uses the method https://api.openai.com/v1/completions.
GPT-3 is a modern language model developed by OpenAI using artificial intelligence. GPT-3 stands for "Generative Pre-trained Transformer 3", which means it uses transformer architecture and pre-training on large volumes of text data. It is capable of generating human-like text, used as a chatbot, creating marketing SEO content, writing code, and debugging it.
To get started, you don't need an OpenAI account. Simply add Albato AI to your automation and set it up to start using it. Albato AI is available on paid Albato plans starting with the Basic plan. To test the functionality, you can use our free OpenAI bot based on this tool.
How to use
IMPORTANT:
The Albato AI tool will deduct 1 transaction for every 250 tokens spent in the response/result. Token count is always rounded up, which means if the OpenAI answer contains 240 tokens, 1 transaction will be deducted, if it contains 270 tokens, 2 transactions will be deducted, and so on. You can view the spent tokens in the Automation Log by clicking on the OpenAI processing result.
A token is a part of a word that is used for natural language processing. For English text, 1 token is approximately 4 characters or 0.75 words. You can read more about tokens in the article — What are tokens and how to count them. You can see usage cases in the "Examples" and "OpenAI Cookbook" sections, and experiment in the "Playground." A token is a part of a word that is used for natural language processing. For English text, 1 token is approximately 4 characters or 0.75 words. You can read more about tokens in the article — What are tokens and how to count them. You can see usage cases in the "Examples" and "OpenAI Cookbook" sections, and experiment in the "Playground".
- Log into your Albato account and head over to the "Automations" section.
- Create a new automation or open an existing one to which you want to add an AI tool.
- Click on the "plus" button to add a new action to the automation.
- Select the "Albato AI" tool and hit the "Add" button.
- In the pop-up window, choose the OpenAI model you want to use and fill in the necessary AI settings for your task.
- Specify the input data for the model in the Promt fields. You can do this by connecting previous automation steps or manually inputting data.
- Run the automation.
Please note that the AI tool is only available on paid Albato plans, so it cannot be used on a trial plan.
Let's go over the available settings:
Model - GPT-3 can understand and generate natural language. Open AI offers four main models with different levels of power suitable for different tasks.
- Davinc is the most capable model
- Ada is the fastest.
The models have different token limits per request. You can read more about it in the Open AI documentation and our previous article.
Prompt - This is the field where you describe the task or request that you want the AI model to perform. It's a free-form text where you can use any words and phrases to describe the task or question you want to ask the AI. A good task description should be clear and specific, providing all the necessary details so that the AI model can perform the task most effectively.
For example, if you want the AI model to create a resume for you, you can use the following description as a prompt:
"Create a resume for me that includes information about my work experience, education, and skills. I have been working in marketing for the past 5 years and have a bachelor's degree in marketing. I would like the resume to be professional and stand out from other candidates. The result should be provided in Word or PDF format.""Create a resume for me that includes information about my work experience, education, and skills. I have been working in marketing for the past 5 years and have a bachelor's degree in marketing. I would like the resume to be professional and stand out from other candidates. The result should be provided in Word or PDF format".
Additionally, it's important to use proper punctuation so that the AI understands whether it needs to perform a specific task or just complete the text. For example, a prompt like "Hello" with a high probability will continue the phrase by answering "Hello, world!", while a prompt like "Hello!" will just answer "Hello!" without any continuation.
Temperature - is a parameter that determines the creativity and diversity of responses generated by AI. It is a numerical coefficient that varies from 0 to 1, where 0 means that the AI will generate responses that are maximally similar to the given prompt, and 1 means that the AI will be more creative and generate responses with greater diversity.
The temperature should be adjusted depending on the task and type of data that the AI is working with. For example, when generating content for a website that requires accuracy and informativeness, a lower temperature value should be used, while a higher temperature value can be selected for generating poetry or fantastic stories.
Suffix - is additional information that can be added to the prompt (the textual description of the task for AI). The suffix provides some context or hint for the AI to better understand what you want to get as a result. For example, we have a prompt: "Write a text for a car sale announcement". Then the suffix may look like this: "Car - Chevrolet Cruze, 2017 model year, mileage 30,000 km, good condition, located in Lisbon". Thus, the suffix contains additional information about the car that needs to be sold, and will help the AI better understand the context of the task and generate a higher quality text of the announcement.
The suffix is not a mandatory parameter, but it can significantly improve the quality and accuracy of the results.
Number of generations - refers to the amount of different answer options that AI will generate based on your Prompt and other parameters such as model and temperature. The more generations you choose, the more options you will receive.
For example, if you set Number of generations to 3, AI will generate 3 different answer options to your Prompt. This can be useful if you want to get multiple ideas or solutions to a problem and choose the most suitable one. However, the more generations you choose, the more time it will take for AI to process your request.
Max tokens - is the maximum number of tokens (text processing units) that AI can use to generate an answer to a given prompt. This is an important parameter to consider when formulating a request to AI. If you specify a value that is too small, AI will not be able to generate complete answers or even ignore your request. On the other hand, if you specify a value that is too large, AI will spend more time processing the request and may produce a response that is too long or unclear.
The maximum value of max tokens depends on the model you are using. For example, for the GPT-3 model from OpenAI, the maximum value of max tokens can be up to 2048. When formulating a request to AI, it is recommended to refer to the values specified in the documentation for the selected model.
