https://app.supportfast.ai/account/?tab=profile&key=true
Supportfast chatbots are powered by OpenAI’s GPT technologies.
By default, the AI model used is GPT-4o-mini, recommended for its excellent balance between quality and cost.
If you want to go beyond usage limits and get unlimited messages with your chatbots, you can connect your personal OpenAI API key. This way, AI usage costs will be charged directly to your OpenAI account, without any limits imposed by Supportfast.
By using your OpenAI key, you also have the option to select a more advanced model than GPT-4o-mini, such as GPT-4o or GPT-4. These models generate higher-quality responses, but they also incur higher usage costs.
Follow this video tutorial (https://www.youtube.com/watch?v=DF8jiO5CY14) to create and use your OpenAI API key. Make sure to enable billing on OpenAI, otherwise your key will not be usable.
Enter your key in the Profile section of your Account. This setting is available only to users with the Owner role.
Once the key is entered, Supportfast will perform a check to verify it is working correctly. If an error is detected, a message in the interface will indicate that the key is not valid. In that case, make sure you have followed all the steps in the video tutorial, including enabling billing on OpenAI.
Once you enter your OpenAI key in the Profile section, all your chatbots will automatically start using it for message processing and training management.
Depending on your subscription, you have access to a limited number of messages without a personal key. These messages have a fixed character limit per response, which may affect the chatbot’s ability to access the knowledge base and provide detailed answers.
When using a personal API key, these limitations are removed:
For this reason, once a valid key is provided, it will always be used with priority over the messages included in your subscription.
In addition to the key set in the profile, you can also assign a different key to each chatbot.
In this case, the key set at the chatbot level takes precedence over the profile key.
This feature allows you to:
By using the OpenAI platform (https://platform.openai.com/), you have the ability to create and manage multiple projects, each with its own dedicated API keys.
This feature is especially useful if you want to clearly separate and monitor the resource usage of your chatbots created with Supportfast. Assigning an API key to a single chatbot is not mandatory, but it is a recommended strategic choice if you want to accurately track the impact and efficiency of each AI assistant.
Within the OpenAI dashboard, you can view a detailed overview for each project, including information such as:
All this data allows you to clearly and objectively analyze the financial performance of your chatbots, helping you optimize budget management and make better decisions about which model to use.
The costs associated with using OpenAI’s APIs depend on several factors: the selected model, the number of enabled tools, and the amount of conversation history stored in context.
In the standard configuration recommended by Supportfast — using the GPT-4o mini model with a context of 5 messages — the estimated cost for approximately 1,000 messages generated by the chatbot ranges between $2.00 and $2.50. This makes GPT-4o mini a particularly advantageous choice for those looking for a good balance between quality, speed, and cost-efficiency.
Of course, you are free to experiment with other configurations by selecting more advanced models or increasing the context size directly from our Playground, allowing you to assess which combination best fits your specific needs.
We strongly recommend keeping an eye on your API usage to avoid unexpected billing charges. You can monitor expenses and usage trends in real time by visiting this link:
https://platform.openai.com/settings/organization/usage
