Setting Up ChatGPT App
Learn how to setup ChatGPT with Picky Assist family of products
In this guide, we will explain how to set up the ChatGPT App within Picky Assist in order to leverage its generative AI capabilities across various Picky Assist products. Previously, ChatGPT was available as an integration through the Connector and Mini App. However, you can now integrate it seamlessly for an improved user experience.
Possibilities with Picky Assist & ChatGPT
Picky Assist has achieved a deep integration with ChatGPT across its platform, enabling businesses to build custom GPT models tailored to specific processes using tools like Mini App Builder and Macro Buttons. Here’s how this integration enhances various Picky Assist products:
Integration with Smart Replies as an AI Chatbot Picky Assist’s smart replies are now powered by ChatGPT, allowing for more natural, improved responses. The key advantage is the ability to train ChatGPT with your own segmented data, categorizing responses to better match customer needs.
Integration with Chatbot Builder When creating WhatsApp chatbots, you can now select the ChatGPT App to handle conversation details. This includes summarizing conversations, extracting valuable information in JSON format, and automatically updating Picky Assist CRM records. The integration streamlines processes like adding events to feeds or pulling key insights from interactions.
Integration with Workflow & Funnels ChatGPT enhances Picky Assist’s workflow and funnels by allowing data-driven decisions. For example, when a CRM status changes, ChatGPT can suggest the next steps for a lead or deal, such as crafting custom sales scripts, personalized emails, or WhatsApp messages, automating the communication process.
Integration with Connectors Picky Assist’s Connectors allow seamless data exchange with third-party platforms. ChatGPT now acts as an intermediary step in these processes, such as automatically extracting relevant details from a job application email, matching it with job descriptions in the CRM, updating candidate records, and sending WhatsApp messages—all without manual intervention.
Integration with Mini App & Macro Button Mini App Builder and Macro Buttons streamline tasks by grouping actions. Businesses can create custom ChatGPT models for specific tasks, such as correcting grammar, training employees with business-specific processes, or summarizing content from uploaded PDFs, enhancing productivity.
Training with Your Own Data One of the standout features of Picky Assist’s integration is the ability to train ChatGPT with your business data at every execution level. This deep integration allows businesses to customize AI responses and actions, making automation more relevant and effective for specific needs.
Limitations & Disclaimer
ChatGPT might not be the optimal solution for every business category. It may not always manage conversations in a fully automated and contextual manner as a human would. Therefore, we recommend monitoring interactions with ChatGPT and always providing users with the option to interact with a human agent.
ChatGPT is not a Free Service
Please remember that ChatGPT and its associated services are not free. OpenAI provides $18 worth of free credit for a period of three months to experiment with their services. After this period, you'll be billed according to your usage. For a clearer understanding of their pricing structure, please refer to their pricing model at https://openai.com/pricing. It's also important to understand how you are billed for the use of tokens. You can find more details in the FAQ section on their pricing page.
You can check your usage and limits at the following link: https://platform.openai.com/account/usage.
Billing by Picky Assist
As of now, we offer ChatGPT as a free app for all plans. However, whenever you send a request to ChatGPT, we consider it as a usage of your Connector Invocation and one credit will be deducted from your available Connector Invocations as per your plan.
Understanding the Basics of ChatGPT
OpenAI, the organization behind ChatGPT, has developed various language models capable of performing a range of tasks. Although OpenAI has developed more than 100 models
GPT-4o: A small language model designed to be a cost-effective alternative, retaining powerful capabilities while being optimized for specific, efficiency-driven tasks.
GPT-4: An advanced model designed for complex tasks requiring deep reasoning and creativity.
GPT-4-turbo: A cost-effective version of GPT-4, optimized for chat applications and offering high performance with reduced costs.
GPT-3.5: A versatile model capable of understanding and generating natural language and code, often used in various applications.
GPT-3.5-turbo: An optimized, cost-effective version of GPT-3.5, ideal for chat applications and traditional completion tasks.
Each model also has different variants, which are useful for our API calls. You can find all the models at https://platform.openai.com/docs/models/overview.
Understanding Prompt & Completion
For the AI to function, we need to provide some form of input known as a "prompt", and the AI returns an output called a "completion".
In simpler terms, the input given to the AI engine is referred to as the "prompt", and the output generated by the AI is known as the "completion".
You can learn more about this at https://platform.openai.com/docs/introduction/prompts-and-completions.
It will also be beneficial to understand the following completion models, which we will utilize in our integration:
Text Completion: https://platform.openai.com/docs/guides/completion/introduction
Chat Completion: https://platform.openai.com/docs/guides/chat
Understanding Tokens
ChatGPT bills you based on tokens, with most pricing models defined such that you pay a certain amount for using 1000 tokens. The cost can vary depending on the model you employ, and ChatGPT bills you separately for the input (prompt) and output (completion).
A useful rule of thumb is that one token usually corresponds to about 4 characters of text in common English usage. This equates to approximately ¾ of a word (meaning that 100 tokens are approximately equal to 75 words).
To estimate the number of tokens you might use, you can employ the Token Calculator at the following link https://platform.openai.com/tokenizer
The more tokens ChatGPT uses, the higher the cost you can expect. To provide you with better control, we allow you to limit the maximum tokens to be used in completion in the default settings of ChatGPT.
Getting Started with ChatGPT
First, we need to get the ChatGPT API token from the OpenAI account, please login into your OpenAI account as shown below; the direct link for https://platform.openai.com/account/api-keys
Click on Create New Secret Key button as shown below and copy your secret key, we need to paste this into the Picky Assist ChatGPT App.
Getting Started with Picky Assist
Login into your Picky Assist account then navigate to Settings -> Apps and find ChatGPT and click on Activate button as shown below;
A popup will appear on the screen where you need to input your Open AI / ChatGPT API Token which we have generated as per the above instruction;
Note of Maximum Number of Tokens Ensure that the token limit you configure is always higher than the number of tokens required for your input. For instance, if you input text that requires 2000 tokens and you set a token limit of 1000, the request will fail. This is because the input itself exceeds the maximum allowed token limit. Thus, it's crucial to estimate the number of tokens your input might use and set your token limit accordingly.
ChatGPT Auto-Disable Scenarios
There are certain situations where our system will automatically disable ChatGPT:
If the provided token is incorrect
If the account is disabled by ChatGPT
If the account does not have an active billing plan
You will also receive an email at your registered address if ChatGPT is disabled under any scenario by our system. In such cases, you will need to manually re-enable it from Settings -> App -> ChatGPT -> Status.
Error Logs
You can view the last 10 error logs to determine the cause if ChatGPT is not functioning as expected.
Picky Assist offers custom, deep enterprise integration of ChatGPT and turnkey solutions for enterprise clients. Read More
Frequently Asked Questions
Why am I getting the error "You exceeded your current quota, please check your plan and billing details. For more information on this error, read the docs" even though I have a ChatGPT Plus or Paid Subscription.
Kindly note that the ChatGPT Plus account and ChatGPT API access accounts are different and you need a separate billing setup for each account, you may set the API account billing through this link https://platform.openai.com/settings/organization/billing/overview , you have the option to add money in advance like a prepaid credit wallet and ChatGPT will debit the account as per the usage of your API. The Billing Interference looks like the below
Last updated
